DBA Cockpit
DBA Cockpit
DBA Cockpit
For SAP systems based on SAP NetWeaver 7.3 SP 9 and higher Enhancement Package 2 for SAP NetWeaver 7.0 SP 4 and higher
Public 09/05/2011
SAP AG Dietmar-Hopp-Allee 16 69190 Walldorf Germany T +49/18 05/34 34 24 F +49/18 05/34 34 20 www.sap.com
Copyright 2011 SAP AG. All rights reserved. No part of this publication may be reproduced or transmitted in any form or for any purpose without the express permission of SAP AG. The information contained herein may be changed without prior notice. Some software products marketed by SAP AG and its distributors contain proprietary software components of other software vendors. Microsoft, Windows, Outlook, and PowerPoint are registered trademarks of Microsoft Corporation. IBM, DB2, DB2 Universal Database, System i, System i5, System p, System p5, System x, System z, System z10, System z9, z10, z9, iSeries, pSeries, xSeries, zSeries, eServer, z/VM, z/OS, i5/OS, S/390, OS/390, OS/400, AS/400, S/390 Parallel Enterprise Server, PowerVM, Power Architecture, POWER6+, POWER6, POWER5+, POWER5, POWER, OpenPower, PowerPC, BatchPipes, BladeCenter, System Storage, GPFS, HACMP, RETAIN, DB2 Connect, RACF, Redbooks, OS/2, Parallel Sysplex, MVS/ESA, AIX, Intelligent Miner, WebSphere, Netfinity, Tivoli and Informix are trademarks or registered trademarks of IBM Corporation. Linux is the registered trademark of Linus Torvalds in the U.S. and other countries. Adobe, the Adobe logo, Acrobat, PostScript, and Reader are either trademarks or registered trademarks of Adobe Systems Incorporated in the United States and/or other countries. Oracle is a registered trademark of Oracle Corporation. UNIX, X/Open, OSF/1, and Motif are registered trademarks of the Open Group. Citrix, ICA, Program Neighborhood, MetaFrame, WinFrame, VideoFrame, and MultiWin are trademarks or registered trademarks of Citrix Systems, Inc. HTML, XML, XHTML and W3C are trademarks or registered trademarks of W3C, World Wide Web Consortium, Massachusetts Institute of Technology. Java is a registered trademark of Sun Microsystems, Inc.
JavaScript is a registered trademark of Sun Microsystems, Inc., used under license for technology invented and implemented by Netscape. SAP, R/3, SAP NetWeaver, Duet, PartnerEdge, ByDesign, Clear Enterprise, SAP BusinessObjects Explorer and other SAP products and services mentioned herein as well as their respective logos are trademarks or registered trademarks of SAP AG in Germany and other countries. Business Objects and the Business Objects logo, BusinessObjects, Crystal Reports, Crystal Decisions, Web Intelligence, Xcelsius, and other Business Objects products and services mentioned herein as well as their respective logos are trademarks or registered trademarks of SAP France in the United States and in other countries. All other product and service names mentioned are the trademarks of their respective companies. Data contained in this document serves informational purposes only. National product specifications may vary. These materials are subject to change without notice. These materials are provided by SAP AG and its affiliated companies ("SAP Group") for informational purposes only, without representation or warranty of any kind, and SAP Group shall not be liable for errors or omissions with respect to the materials. The only warranties for SAP Group products and services are those that are set forth in the express warranty statements accompanying such products and services, if any. Nothing herein should be construed as constituting an additional warranty. Disclaimer Some components of this product are based on Java. Any code change in these components may cause unpredictable and severe malfunctions and is therefore expressively prohibited, as is any decompilation of these components. Any Java Source Code delivered with this product is only to be used by SAPs Support Services and may not be modified or altered in any way. Documentation in the SAP Service Marketplace You can find this documentation at the following address:
http://service.sap.com/instguides
Typographic Conventions
Type Style Example Text Description Words or characters quoted from the screen. These include field names, screen titles, pushbuttons labels, menu names, menu paths, and menu options. Cross-references to other documentation Example text Emphasized words or phrases in body text, graphic titles, and table titles Technical names of system objects. These include report names, program names, transaction codes, table names, and key concepts of a programming language when they are surrounded by body text, for example, SELECT and INCLUDE. Output on the screen. This includes file and directory names and their paths, messages, names of variables and parameters, source text, and names of installation, upgrade and database tools. Exact user entry. These are words or characters that you enter in the system exactly as they appear in the documentation. Variable user entry. Angle brackets indicate that you replace these words and characters with appropriate entries to make entries in the system. Keys on the keyboard, for example, F2 or ENTER.
Icons
Icon Meaning Caution Example Note Recommendation Syntax
EXAMPLE TEXT
Additional icons are used in SAP Library documentation to help you identify different types of information at a glance. For more information, see Help on Help General Information Classes and Information Classes for Business Information Warehouse on the first page of any version of SAP Library.
Example text
Example text
<Example text>
EXAMPLE TEXT
Database Administration Using the DBA Cockpit: IBM DB2 for LUW
Contents
Database Administration Using the DBA Cockpit: IBM DB2 for Linux, UNIX, and Windows ................................................................ 9
1 Introduction to the DBA Cockpit ..................................................... 10
1.1 Web Browser-Based User Interface (Web Dynpro) ........................... 12 1.2 Maintenance Actions in the DBA Cockpit .......................................... 17 1.3 SAP GUI-Based User Interface (Deprecated) ..................................... 20
2.2 Enabling the Database for the Data Collection Framework .............. 32
2.2.1 Setting Up the Data Collection Framework (DCF) Manually .................................... 34 2.2.2 Maintaining Templates for the DCF......................................................................... 35
2.3 Management of the System Landscape Using the Central Calendar ...................................................................................................... 39
2.3.1 Using the Central Calendar .................................................................................... 40
3.2 SQL Command Line ............................................................................. 50 3.3 The Index Advisor ................................................................................ 50
3.3.1 Retrieving Index Recommendations for the Dynamic SQL Cache ........................... 51 3.3.2 Retrieving Index Recommendations for a Single SQL Statement ............................ 53 3.3.3 Defining Virtual User-Defined Indexes .................................................................... 56 3.3.4 Validating Indexes Using the EXPLAIN Function .................................................... 58 3.3.5 Creating Indexes in the ABAP Dictionary ................................................................ 58
4 Performance ...................................................................................... 60
4.1 Performance: Performance Warehouse ............................................. 60
4.1.1 Performance Warehouse: Reporting....................................................................... 61 4.1.2 Performance Warehouse: Configuration ................................................................. 62
4.2 Performance: Time Spent Analysis .................................................... 65 4.3 Performance: Workload Statistics ...................................................... 71 4.4 Performance: Top SQL Statement Analysis ...................................... 73 4.5 Performance: Snapshots ..................................................................... 78
4.5.1 Snapshots: Database ............................................................................................. 80 4.5.1.1 Displaying Detailed Database Snapshot Data ................................................... 82 4.5.2 Snapshots: Schemas.............................................................................................. 83 4.5.3 Snapshots: Buffer Pools ......................................................................................... 84 4.5.3.1 Displaying Detailed Buffer Pool Snapshot Data ................................................. 86
September 2011
4.5.4 Snapshots: Tablespaces ........................................................................................ 87 4.5.4.1 Displaying Detailed Tablespace Snapshot Data ................................................ 89 4.5.5 Snapshots: Containers (Only DB2 V9.7 or Higher) .................................................. 89 4.5.6 Snapshots: Tables.................................................................................................. 91 4.5.7 Snapshots: Applications (Only DB2 V9.7 or Higher) ................................................ 94 4.5.8 Snapshots: Applications (Only DB2 V9.5 or Lower)............................................... 100 4.5.9 Snapshots: SQL Cache (Only DB2 V9.7 or Higher)............................................... 102 4.5.10 Snapshots: SQL Cache (Only DB2 V9.5 or Lower) ............................................. 105 4.5.11 Snapshots: System............................................................................................. 110
4.8 History: Database ............................................................................... 122 4.9 History: Tables ................................................................................... 124
5.5 Space: Containers .............................................................................. 141 5.6 Space: File Systems........................................................................... 142 5.7 Space: Tables and Indexes................................................................ 144
5.7.1 Tables and Indexes: Top Space Consumers......................................................... 145 5.7.2 Tables and Indexes: REORG Candidates ............................................................. 146 5.7.3 Tables and Indexes: Compression Status ............................................................. 147 5.7.4 Tables and Indexes: Virtual Tables ....................................................................... 151 5.7.5 Tables and Indexes: Indexes ................................................................................ 152
September 2011
Database Administration Using the DBA Cockpit: IBM DB2 for LUW
7.2 Configuration: Database Manager .................................................... 186 7.3 Configuration: Database.................................................................... 187
7.3.1 Maintaining the Database Configuration ............................................................... 189 7.3.2 Comparing Database Configuration Parameters for Several Database Partitions .. 189
7.4 Configuration: Registry Variables .................................................... 190 7.5 Configuration: Parameter Check ...................................................... 190 7.6 Configuration: Configuration History ............................................... 195 7.7 Configuration: Buffer Pools .............................................................. 197
7.7.1 Maintaining Buffer Pools ....................................................................................... 199
7.10 Configuration: Special Tables Regarding RUNSTATS.................. 211 7.11 Configuring Automatic Maintenance Functions ............................ 213
7.11.1 Automatic Maintenance Settings: General Settings ............................................. 216 7.11.2 Automatic Maintenance Settings: Automatic Backup ........................................... 217 7.11.3 Automatic Maintenance Settings: Automatic RUNSTATS.................................... 220 7.11.4 Automatic Maintenance Settings: Automatic REORG.......................................... 221
8.2 The DBA Log....................................................................................... 257 8.3 Back-End Configuration .................................................................... 258 8.4 The SQL Script Maintenance ............................................................. 259
September 2011
10.7 Diagnostics: Trace Status ............................................................... 280 10.8 Diagnostics: Cumulative SQL Trace............................................... 282
September 2011
Database Administration Using the DBA Cockpit: IBM DB2 for LUW
12.17 Tablespace Details: Direct I/O ....................................................... 316 12.18 Tablespace Details: XML Storage ................................................. 316 12.19 Application Details: Application ................................................... 317 12.20 Application Details: Agents ........................................................... 321 12.21 Application Details: Assigned Agents .......................................... 323 12.22 Application Details: Agents Memory ............................................ 325 12.23 Application Details: Buffer Pool.................................................... 325 12.24 Application Details: Direct I/O ....................................................... 327 12.25 Application Details: XML Storage ................................................. 327 12.26 Application Details: Locks and Deadlocks .................................. 329 12.27 Application Details: Calls .............................................................. 330 12.28 Application Details: Sorts .............................................................. 331 12.29 Application Details: Cache ............................................................ 332 12.30 Application Details: Unit of Work.................................................. 333 12.31 Application Details: Statement...................................................... 333 12.32 Application Details: Statement Text ............................................. 335 12.33 Application Details: SQL Workspace ........................................... 337
September 2011
Database Administration Using the DBA Cockpit: IBM DB2 for Linux, UNIX, and Windows
Use
This document explains how you administer your database using the DBA Cockpit that allows you to monitor, control, and configure your database. The DBA Cockpit provides you with access to all the functions and indicators for monitoring and administration: Checking system status and operation modes Locating potential problems as quickly as possible Early diagnosis of potential problems, for example, resource problems in the host or database system, which could adversely affect the SAP system Analyzing and tuning the SAP system and environment (host and database systems) to optimize the throughput of the SAP system Configuring the database
Naming Conventions
Throughout this document, the following naming conventions apply: IBM DB2 Version 9.7 for Linux, UNIX, and Windows is referred to as DB2 V9.7. IBM DB2 Version 9.5 for Linux, UNIX, and Windows is referred to as DB2 V9.5. IBM DB2 Version 9.1 for Linux, UNIX, and Windows is referred to as DB2 V9.1.
More Information
For additional information about running an SAP system on DB2 for Linux, UNIX, and Windows, choose SAP on DB2 for LUW in the SDN in the Useful Links section of the DBA Cockpit. For DB2-specific information, see the respective IBM DB2 Information Center and IBM manuals for your database at: https://www-04.ibm.com/support/docview.wss?rs=71&uid=swg27009474
September 2011
10
September 2011
Example All BW-specific functions of the DBA Cockpit require additional BW-functionality in the back end. End of the example. The connections that are assigned to the system, such as the availability of RFC connections The user permissions, which are SAP-specific and also database-specific For more information, see Maintenance Actions in the DBA Cockpit [page 17]. The configuration of the database, such as monitoring settings or automatic maintenance settings The configuration of the back end For more information, see Enabling the Database for the Data Collection Framework [page 32]. Depending on conditions the currently selected system is fulfilling, actions might either be disabled or not visible in the DBA Cockpit. In addition, these actions might show a reduced amount of information on screens and specific functions of affected actions might be disabled (for example, administrative functions are not available due to missing permissions).
More Information
Web Browser-Based User Interface [page 12] SAP GUI-Based User Interface (Deprecated) [page 20]
September 2011
11
Favorites list
12
September 2011
Navigation and Screen Layout of the Web Browser-Based User Interface Area Description
Common header Provides a standard set of functions, for example, to log off from the DBA area Cockpit or to customize the layout. In the top level navigation, you can switch between the following areas: Cross-system area on the System Landscape tab page Provides information about the overall system landscape Database-specific area on the Database tab page Provides information about the selected database Top level navigation including second-level navigation In the second-level navigation, the main task areas of database administration are provided, for example, performance monitoring, space management, and job scheduling. For fast navigation, these main task areas provide pull-down menus corresponding to the related detail levels. You can hide the areas Detail Navigation, System Landscape Selector, and Favorites by choosing the Expand or Collapse Launchpad pushbutton on the left side of the top level navigation area. If the launchpad is collapsed, the second-level navigation part provides a simplified system selection field with an F4 help. Your chosen screen layout is stored in the user settings and restored at the next start of the DBA Cockpit. Contains the main actions of the main task areas. Depending on the selected main action, a subset of related actions is available. Example Detail navigation If you choose Performance in the top level navigation area, the main actions become available, such as Performance Warehouse, Time Spent Analysis, and others. If you choose Performance Warehouse, the subactions Reporting and Configuration become available. End of the example. System landscape selector Provides a quick overview of all configured systems. This area is described in more detail under Customizing of the System Landscape Selector later in this section.
September 2011
13
Contains a list of favorite links to special tools and actions. To provide quick access to specific tools and actions, choose Personalize Add Favorite in the common header area. An entry is added to your list of favorites. You can rename or delete favorites by choosing Personalize Organize Favorites . For more information, see Special Tools in the Favorites List [page 43]. Contains the following useful links to: The IBM DB2 Information Center Useful links By choosing this link, you can directly access online information about DB2. SAP on DB2 for LUW in the SAP Developer Network (SDN) By choosing this link, you can directly access information about the DB2 SAP community in the SDN, such as, articles and blogs. Displays the message window that is provided by the framework. Unlike the classic SAP GUI message processing, this window contains a complete history of all messages that are sent during the session. In addition, you can: Collapse or expand the window by choosing Expand Message Window or Collapse Message Window. Check if a long text for a message is available by double-clicking the message or by choosing Details. Note By default, the message window is collapsed. When a new message is generated, it is automatically expanded. End of the note. Global toolbar The global toolbar provides a set of globally available functions for navigation and content-related functions like Refresh.
Favorites list
Central system This area is common to most actions providing, for example, the time of the data last refresh, the startup time, and the database name.
14
September 2011
Displays details of the currently selected action. The content area is divided into the following areas that are optionally available depending on the chosen action: The Selection area where you can enter selection criteria for the content to be displayed Content area The Summary area that provides summary views of data, for example, totals or execution times The content, which depends on the screen and action you have chosen You can refresh the content by choosing the Refresh pushbutton in the global toolbar or by changing the selection criteria and then choosing the Apply Selection pushbutton in the Selection area. Some reports are displayed as a chart. To modify the chart view, you can use the Chart Type, Values, and Chart Size menu buttons that are optionally available depending on the screen. By choosing the Chart Size menu button, you can specify the width and height of the chart to be displayed. In addition, you can open a detailed table view by choosing the Toggle Table Display pushbutton. Content detail area Only appears with certain actions and displays additional information that is related to the selected action. Typically, this area shows details that are related to the information provided in the main content area.
To use a custom grouping, you must first define and add a custom group to the list. To do so, choose Add Group from the pop-up menu of the menu button Group Systems by Selected Criteria. Specify a name for the custom group and assign the systems of your choice.
September 2011
15
As soon as you have added a custom group, the option Organize Groups becomes available in the pop-up menu of the menu button Group Systems by Selected Criteria, which lets you maintain an already existing group. Filter Systems by Selected Criteria You can filter the list of available systems to show only those systems that match the filter criteria. You filter, for example, by the alert status of the systems. Search Systems Provides an input field where you can search for a specific system in the list.
Switching from the Web Browser-Based User Interface (Web Dynpro) to the SAP GUI-Based User Interface
You have to switch back to the SAP GUI-based user interface of the DBA Cockpit in the following cases: The system to be monitored is using DB2 UDB Version 8.2 and there is specific information for DB2 UDB Version 8.2 that is not available in the Web browser-based user interface. In a heterogenous system landscape, you select a database to be monitored that does not yet fully support the Web browser-based user interface To switch the user interface, perform the following steps: 1. In your SAP system, you enter transaction DBACOCKPIT. Depending on your settings, the DBA Cockpit either starts with the Web browserbased user interface in embedded mode or with the SAP GUI-based user interface. Note Embedded mode means that the Web browser-based user interface does not open in a separate Web browser but is embedded in the SAP GUI. End of the note. 2. You switch between the two user interfaces as follows: o o To switch from the Web browser-based user interface to the SAP GUI-based user interface, choose the Switch to SAPGUI menu button To switch from the SAP GUI-based user interface to the Web browser-based user interface, choose the Switch to Web Dynpro pushbutton. Note To open the Web browser-based user interface in the full screen mode, choose the Full Screen option of the menu button. End of the note. If you want the SAPGUI to be the default UI to start with, choose the option Use SAP GUI as default UI of the Customize button in the SAP GUI. To set the WebDynpro UI as default, choose the option Use Web Dynpro as default UI.
16
September 2011
September 2011
17
The attributes DBA_DBHOST, DBA_DBSID, and DBA_DBUSER must match the corresponding attributes for the database connection that was assigned to the selected system. The special value <LOCAL SYSTEM> for the attribute DBA_DBSID is used to identify the local system itself. The attribute ACTVT of this S_DBCON authorization object defines the level of permitted actions and can have the following values: Value 03 Display Description Enables read access to all screens of the DBA Cockpit except to those that only have a maintenance mode and no read-only mode Enables read and maintenance access to all screens of the DBA Cockpit except to those that require extended maintenance permissions Enables read and maintenance access to all screens of the DBA Cockpit including special maintenance screens 36 Extended maintenance Note The only screen for which extended maintenance permission is required is the SQL Command Line screen that you can access in the Favorites list of the DBA Cockpit. End of the note. You can grant authorizations for using the DBA Cockpit with the following roles: SAP_BC_S_DBCON_USER Read-only role that allows monitoring access to all systems configured within the DBA Cockpit SAP_BC_S_DBCON_ADMIN Additionally grants administration rights to the user for all systems. This role does not include the value Extended maintenance. Note Make sure that you have maintained the authorizations for your DBA user and for all background users that either run jobs of the DBA Planning Calendar or the SAP standard jobs SAP_COLLECTOR_FOR_PERFMONITOR and SAP_CCMS_MONI_BATCH_DP. End of the note.
23 Maintain
18
September 2011
Locking of Actions
For each maintenance action that you have selected using the DBA Cockpit, a lock is set for the system that is being monitored. All locks are released when you exit the DBA Cockpit or when you change to another system.
September 2011
19
History
History
Snapshots
Tables
if DCF
Space
Tables and
In the submenu, you can choose between the following views: Space Tables and Indexes Top Space Consumers REORG Candidates Compression Status Virtual Tables Space History Overview Available but deprecated
Database
Replaced by
Space
Database
Tables
Replaced by Space Tables and Indexes Consumers if DCF is installed Available but deprecated
Configuration Settings
Monitoring To configure the DCF, choose Configuration Data Collection Framework Collector Configuration .
20
September 2011
Parameter
Configuration History
If you want to use the SAP GUI-based user interface, you can change the default user interface as follows: 1. Call transaction DBACOCKPIT. The SAP GUI-based user interface appears and in parallel, the Web browser-based user interface in a separate browser window. 2. In the SAP GUI, choose the Personal Settings menu button and select Use SAPGUI for DB2 LUW Systems. The changes take effect with the next call of transaction DBACOCKPIT. End of the note. For more information about the functions provided on the SAP GUI screens, see Database Administration Using the DBA Cockpit: IBM DB2 for Linux, UNIX, and Windows (For SAP Systems based on Enhancement Package 1 + 2 of SAP NetWeaver 7.0) at: http://service.sap.com/instguidesnw70 Operations Database-Specific Guides
September 2011
21
22
September 2011
Cockpit uses the RFC connection in parallel to the database connection for the same system. Note You can only maintain RFC connections with transaction SM59, not with the DBA Cockpit. End of the note.
Prerequisites
The system(s) that you want to monitor must have a database release that is compatible with the database release of your local database. The user for the database connection must have sufficient database permissions. For more information, see Maintenance Actions in the DBA Cockpit [page 17].
Procedure
1. Call the DBA Cockpit. The System Configuration screen appears displaying a list of all available systems with a Configuration Status icon that indicates the current system status. Note When you start the DBA Cockpit for the first time, the local system is automatically added to the list of all available systems. At least one system entry is displayed. End of the note. 2. Choose the Add pushbutton. The wizard Integrate a System appears.
September 2011
23
The following table lists the steps and recommended actions: Step Name Action Provides an overview of the configuration steps Introduction No action to be taken Specify the following: o Name of the system that you want to monitor Note This name is a unique ID and does not have to be the SAP system ID. You can choose any name except the SAP system ID of the local system that is reserved for the local system entry. End of the note. System Characteristics o o Description of the monitored system Connection type Choose between the following connection types: Remote Database Connection Remote Database Connection and RFC Destination Remote Database Connection via RFC Destination Specify the name of the RFC destination to be used RFC Destination (Optional step that is only necessary if, in the previous step, you have chosen a connection type that requires an RFC destination.) Note The specified RFC destination must already exist and be available. End of the note. You can test the destination by choosing the Test Connection pushbutton. If the connection is not working, you can only proceed if you select the Ignore Communication Errors checkbox.
24
September 2011
Specify the following: o Name of the database connection This is a unique name that you can freely choose. Alternatively, you can search for an already existing connection name using the related search help. In this case, confirm the selected name using the Enter key to prefill the connection attributes. o o Database Connection o Database platform of the monitored database Connection maximum that limits the parallel use of this connection by the kernel Connection optimum that sets the optimum number of open connections Recommendation You should set this value to 0 to ensure that the kernel closes connections if they are no longer used. End of the recommendation. o o User name and password for the user that is used for monitoring List of connection parameters that are required to identify the database
Specify how you want to collect monitoring data: o If alerts are to be provided for the RZ20 alert monitor, select the Activate Alert Monitoring checkbox. Note If you are using the DBA Cockpit in SAP Solution Manager 7.1, this option should not be used anymore. Instead, the E2E alerting of SAP Solution Manager is used that requires no specific setup in the DBA Cockpit. End of the note. o o If the monitoring data is to be collected by the remote system, select the Data Collection by Remote System checkbox. If data for the Central Planning Calendar is to be provided, select the Show Scheduled Jobs in Central Planning Calendar checkbox. Deprecated option: If data about the performance or the size of database objects is to be collected, select the Collect Space and Performance History Data
Monitoring Settings
September 2011
25
checkbox. Note This option has been deprecated and should not be used anymore. Instead, proceed as described in Enabling the Database for the Data Collection Framework [page 32]. End of the note. Depending on the selected database platform, only a subset of options might be available. Summarizes all actions to be performed Summary To save your entries, choose the Execute pushbutton. Summarizes all performed actions including error messages Execution Protocol To exit the wizard, choose the Finish pushbutton and return to the System Configuration screen.
3. Optional: If you want to change an existing configuration entry, select the system entry in the overview list and choose the Change pushbutton. In the dialog box Change System Configuration Entry, enter and save your changes. 4. Optional: If you want to delete a configuration entry, select the system entry in the overview list, choose the Delete pushbutton and confirm the deletion.
2.1.2 Configuring Systems for Remote Monitoring Using the System Landscape Directory (SLD)
The system landscape directory (SLD) contains data from all database systems that are available in your system landscape. You can use this data to set up the system configuration in the DBA Cockpit instead of setting it up manually or using SAP Solution Manager. When you set up the DBA Cockpit for the first time, you use this procedure to import the appropriate data from the SLD. During production operation, you use the procedure to synchronize the data between the SLD and the DBA Cockpit periodically. Note This function has been deprecated. We recommend that you only use the SAP Solution Manager for the setup and monitoring of your system landscape. End of the note.
26
September 2011
Procedure
1. To import database connection data from the SLD, call the DBA Cockpit. 2. On the System Landscape tab page, choose SLD System Import. The SLD System Import screen appears. Depending on the system landscape, one or more of the following nodes are displayed: o New Database Systems in the SLD All database systems that are registered in the SLD and that so far have been unknown to the DBA Cockpit are displayed. o Changed Systems From Earlier SLD Imports All database systems for which the main data differs between the SLD and the DBA Cockpit are displayed. o Systems No Longer Registered in the SLD All systems that were originally imported from the SLD into the DBA Cockpit but that are no longer registered in the SLD are displayed. o Systems Identical in the SLD and in the DBA Cockpit All systems that are registered in the SLD and that are identical in the DBA Cockpit are displayed. o Unsupported Database Systems in the SLD All database systems that are registered in the SLD but not supported by the DBA Cockpit are displayed. Note Each database system is described as follows: <Name (system ID) of the database system> on <main database host> ( <database platform> ) End of the note. 3. To import database system data, choose the Change pushbutton. The actions allowed for each database system are displayed in the second column of the tree. 4. Select the actions that you want to execute for the selected database systems and choose the Import pushbutton. By default, only the import of new database systems is selected.
September 2011
27
The selected actions are executed. A short message for each executed action is displayed in the content detail area. Note Connection data that is retrieved from the SLD might not be complete for one of the following reasons: o o Depending on the data provided by a system to the SLD, some connection data can be incomplete. User or password data is not available via SLD.
When you establish the connection to an imported system for the first time, the DBA Cockpit checks the completeness of the configured system. This means that you are prompted for user, password, and connection information, if necessary. If additional connection information is required, enter the required data as described in Configuring Database Connections [page 28]. End of the note.
Procedure
1. Call the DBA Cockpit and on the System Landscape tab page, choose Database Connections. The Database Connections screen appears displaying a list of all available database connection definitions grouped by database platform: Column Description Name of the database connection Remote Database Connection Note This is a unique name that you can freely choose. End of the note. DB Name Name of the database
28
September 2011
Column DB Host DB Schema User Name Permanent Max. Connections Opt. Connections
Description Name of the database host Name of the database schema to be monitored Name of the connect user Specifies whether the connect user must be permanently available Maximum allowed number of open connections Optimal number of connections
By default, the database connections that are defined in the local system are displayed. 2. To add a database connection, choose the Add pushbutton. The Add Connection Entry dialog box appears. 3. In the Add Connection Entry dialog box, enter the following: o In the Connection Name field, specify the name of the connection. Note This name is a unique ID that you can choose freely except for names that are reserved by SAP for generated connections. These can be, for example, administrator connections or connections that are used by systems from the system landscape directory (SLD). End of the note. o o In the Database System field, select the name of the database platform from the dropdown list. In the Connection Maximum field, enter an appropriate value. This value limits the number of database connections that are currently held by the SAP system. The SAP system does not let you exceed this limit. In the Connection Optimum field, enter an appropriate value. This value is a more flexible limit that can be exceeded. If you want the connection to be mandatory for the SAP system, select the Permanent checkbox. This parameter defines the availability of the database connection. It is then handled in the same way as the local default connection, that is, if this database connection is not available for a work process, the work process of the SAP system cannot run. Caution You should set this parameter only if the connection is absolutely required to run your SAP system.
September 2011
29
End of the caution. o In the User Name field, enter the name of the connect user. Make sure that you choose a user with the appropriate authorizations. For more information, see Maintenance Actions in the DBA Cockpit [page 17]. In the Password field, enter a password for the connect user and confirm it in the appropriate field. In the parameter table, specify the following additional database-specific parameter values: Attribute Database Name Name of the database Name or number of the service Service Name This value corresponds to the parameter SVCENAME of the database manager configuration (DBM) of the remote database. Name of the remote database server Name of the schema to be monitored Schema Name Note If you omit this field, the name of the SAP connect user is used as schema. End of the note. 4. To save your entries, choose the Add pushbutton. 5. To change a database connection, expand the respective database platform in the Remote Database Connection column, select a connection, and choose the Change pushbutton. The Change Connection Entry dialog box appears. 6. Enter your changes in the appropriate fields and choose Save. 7. To check if you have entered the correct user and password information as well as the correct technical connection data, you can test a database connection. To do so, select a connection and choose the Test pushbutton. The test result is displayed in the Connection Test Protocol in the content detail area. 8. To delete a connection, select a connection and choose the Delete pushbutton. Caution If the selected database connection is still used by a system that is registered in the DBA Cockpit, you cannot delete it. End of the caution. Description
o o
Database Host
30
September 2011
Handle
September 2011
31
Column Reconnect
Force Reconnect Indicates if the connection is forced to reconnect after it was closed Timeout Maximum Optimum Date Time DB Host Program Timeout when the connection is closed Maximum allowed number of open connections Optimal number of connections Date when the connection was established Time when the connection was established Host name Program that opened the connection
For more information about a database connection, select the relevant entry from the overview table. The Database Connection Details area appears displaying details about the database connection, such as technical settings, user credentials, and parameter values.
32
September 2011
If the DCF does not yet exist, it is automatically set up if the following applies: The minimum database release level of the monitored database is at least DB2 V9.1. The database is enabled for automatic storage management or the required tablespaces have already been created. If the DCF cannot automatically be set up because of missing tablespaces or insufficient privileges of the monitoring user, the DBA Cockpit initially displays the appropriate configuration screen where you can perform the required setup steps as described in Setting Up the Data Collection Framework (DCF) Manually [page 34]. All settings for the instrumentation, for example, recurrence patterns for the scheduled data collectors and the amount of data to be collected are determined by templates. SAP provides the following standard templates: SAP Default Most of the data collectors run hourly. The sliding window is set to two weeks except for the most important tables and indexes. SAP Detailed Most of the data collectors run quarter-hourly. The sliding window is set to two weeks except for the most important tables and indexes. By modifying these templates, you can influence the automatic setup of the DCF. For more information, see Maintaining Templates for the DCF [page 35]. Note If you set up the data collection framework, the monitoring history function that has been deprecated as of Enhancement Package 2 for SAP NetWeaver 7.0 is switched off, that is: The option Collect Space and Performance History Data in the system configuration entry is disabled so that the related SAP background jobs do not collect history data anymore for SAP releases prior to Enhancement Package 2 for SAP NetWeaver 7.0. If the REORGCHK for All Tables job is scheduled, it is removed from in the DBA Planning Calendar and the respective data is now collected by the DCF. The special compression function that formerly was provided by the REORGCHK for All Tables job, has been integrated in the Compression Status screen that is available under Space Tables and Indexes . End of the note.
September 2011
33
Procedure
1. Call the DBA Cockpit, and on the Database tab page, choose Collection Framework Collector Configuration . The Collector Configuration screen appears. 2. Choose the Implement Template pushbutton. The Implement Template dialog box appears. The default template is marked with an asterisk (*). By default, deprecated functions are preselected for disabling. If you want to keep any old settings, deselect the relevant checkboxes. Note The option to disable deprecated functions or to keep them is only available if there are deprecated functions that are currently active but recommended for disabling. End of the note. 3. To set up the template in the monitored database, choose the Implement pushbutton. Configuration Data
More Information
Configuration: Data Collection Framework [page 227]
34
September 2011
September 2011
35
Column
Description
Release (min) Minimum database release required for the history data collector Release (max) Maximum database release supported by the history data collector Indicates the setup status: Green: The history data collector is set up as intended. Yellow: The history data collector is not set up properly. We recommend that you repair it. Red: The history data collector is not set up. Set up the history data collector according to the assigned template. Gray: The history data collector is not set up and the assigned template also marks this data collector to be deactivated. Collection Interval Schedule of the history data collector State of the DB2 event monitor: Green: The DB2 event monitor is set up as intended. Yellow: Event Monitor Active The DB2 event monitor is not set up properly. We recommend that you repair the history data collector. Red: The DB2 event monitor is not set up. Set up the data collector according to the assigned template. Gray: No DB2 event monitor is used by the data collector. Days Kept in History Number of days of the sliding history window The data in the history is deleted after this number of days.
Setup
36
September 2011
2. On the Collector Configuration screen, choose the Save as New Template pushbutton. Note The Save As New Template pushbutton is only enabled if the respective system configuration fulfills the minimum template requirements of SAP that are defined by the SAP Default template. This means that every new template must determine a finer granularity or a longer retention time for the history data than the SAP Default template does. End of the note. 3. In the Save As New Template dialog box, enter a unique name and a description of the new template. 4. Save the template.
September 2011
37
Note If you want to discard your changes, choose the Reset Changes pushbutton. End of the note. In some areas, the DBA Cockpit joins data from different data collectors on the time dimension. We, therefore, recommend that you have the same configuration for the schedules for these groups of data collectors: General data collectors o o o o o Buffer Pools Containers Configuration Database Tablespaces
Deleting a Template
You can delete a template by choosing it from the Template dropdown list and choosing the Delete pushbutton. A list of all systems is displayed that are assigned to this template. If there are entries in the list, the default template is assigned to these systems before the template is deleted. Note You cannot delete SAP templates or the current default template. If you want to delete the current default template, you first have to unmark it as the default by setting a different template as the new default.
38
September 2011
Features
You can manage in real time systems directly administered from the system where the DBA Cockpit is running as well as remote systems, including non-ABAP systems. You can quickly check the color-coded status for each system to see if actions have executed successfully. You can quickly check the number of actions and number of actions with the highest status severity for each system, as shown in the following example: Example The entry for February 2007 in the cell for Thursday February, 08 is: 3 FUD 2 On system FUD for Thursday 8th January 2007 (in the past), there were 3 actions planned, 2 of which had the highest status severity. For example, the highest status severity might be Warning, in which case the entry is displayed with a yellow background. End of the example.
September 2011
39
Activities
1. On the system where you normally run the DBA Cockpit, you plan a regular job in the DBA Planning Calendar to update the results from remote systems using the action Central Calendar Log Collector. For example, you plan this job to run daily at 06:00. 2. You define the systems you want to monitor in the DBA Cockpit by setting the flag Collect Central Planning Calendar Data for each system. 3. You regularly check the results using the Central Calendar. 4. If you need to schedule, change, delete, or execute actions, you switch to the DBA Planning Calendar.
Prerequisites
You have defined the systems to be displayed in the Central Calendar by selecting the required system on the System Configuration screen on the System Landscape tab page of the DBA Cockpit and selecting the Collect Central Planning Calendar Data checkbox under Collector Options in the System Configuration Details area. For more information, see Configuring Systems for Remote Monitoring Using Remote Database Connections. In an ABAP system, make sure that you schedule the jobs for the remote database in the central monitoring system. Jobs that have been scheduled in the remote system are not displayed. In the DBA Planning Calendar of the DBA Cockpit where you call the Central Calendar, you have planned the action Central Calendar Log Collector to run regularly. This collects information from the defined remote systems for display in the Central Calendar. For more information, see Setting Up and Working with the DBA Planning Calendar [page 236].
Procedure
1. Start the Central Calendar by choosing Central Calendar on the System Landscape tab page of the DBA Cockpit. The Central Calendar is displayed. If you have already run or planned actions, you see entries by day, one for each system, which provide a summary of the scheduled jobs. The following is an example of entries for Thursday February, 08 affecting two systems, FUD and FIB: Example 3 4 FUD FIB 2 1
40
September 2011
On system FUD for Thursday, February 8th, there were three actions planned, two of which had the highest status severity. For example, the highest status severity for FUD might be Finished with warning, in which case the entry for FUD is displayed with a yellow background. This means that two actions ended with a warning. On system FIB for the same day, there were four actions planned, one of which ended with the highest severity. For example, the highest severity for FIB might be Finished with error, in which case the entry for FIB is displayed with a red background. This means that one action ended with an error. The following table shows the color-coded statuses in the Central Calendar, which you can also see by choosing the Legend pushbutton: Color Light blue Dark blue Green Yellow Red Dark yellow Dark red End of the example. 2. For a summary of the actions for a day, click the relevant day header. A summary of the actions and the status for each system on the day you selected are displayed in the Details area, as in the following example: System Total FUD FIB 3 4 No longer available 1 3 Scheduled Running Finished Warning Error Overdue 2 1 Not yet available Planned Running Finished successfully Finished with warning Finished with error No longer available Scheduling failed Status
3. To obtain new statistics of the scheduled jobs for all systems, choose the Refresh in the Background pushbutton. 4. In the Schedule Refresh dialog box, you can choose between the following options: Option Run in Dialog Start Immediately in the Description Refresh runs in dialog mode, which can take a long time. Therefore, we usually do not recommend this option. Refresh runs immediately as a background job
September 2011
41
Description
Refresh runs in the background on the date and at the time that you specify
6. We recommend that you schedule the action Central Calendar Log Collector to run regularly as described above in Prerequisites. 7. End of the recommendation. 8. To refresh the display with the newly gathered statistics of the previous step, choose the Refresh pushbutton. 9. Optional step: In the Selection area, you can customize the calendar display as follows: o o Specify dates or time periods to be displayed Specify a factory calendar Holidays are displayed in the same background color as weekend days. This does not restrict the planning of actions in the DBA Planning Calendar.
42
September 2011
September 2011
43
Note The statements might contain optional comments such as --OPTLEVEL( <optlevel> ) -QUERY_DEGREE(< query_degree> --LOCATION( <report> , <position> ). If no comments are specified, the statements are explained using the default <optlevel> and the default <query_degree> for the work process. End of the note. If a statement was explained successfully, information about the SQL statement text is provided on the following tab pages: Tab Page Original Statement Optimized Statement Access Plan Description Displays the original SQL statement
Displays the SQL statement that was rewritten by the DB2 optimizer Displays the access plan that was generated by the DB2 optimizer Displays the output of the EXPLAIN snapshot
EXPLAIN Snapshot
Note The EXPLAIN Snapshot tab page is only available if the monitored database is DB2 V9.5 for Linux, UNIX, and Windows or higher. End of the note.
44
September 2011
Catalog Information (Optional) Displays details for the respective catalog object of the selected operator
Search for operators in a complex statement by choosing the Find pushbutton Open an extra navigation window for complex access plans by choosing the Navigate pushbutton Print the graphic by choosing the Print pushbutton Configure the graphic before you print it by choosing theStandard Graph or Detailed Graph pushbutton Display or hide all operators by choosing Collapse or Expand Display global details about the access plan by choosing theToggle pushbutton Display information about the JNet version used (can be required by SAP Support) by choosing the About pushbutton Note For each index used in the access plan, the number of key columns that were really used within the access plan is displayed. In the appropriate tool tip, the used index field names are also displayed. Volatile tables and indexes of volatile tables are marked with an extra volatile label. To change and re-explain the SQL statement, choose Edit Statement. End of the note.
September 2011
45
Call transaction ST05 and choose Enter SQL statement. Enter an SQL statement manually and choose Explain. If a statement cannot be explained, the ERROR: Check SQL Statement screen appears providing a detailed error message and the possibility to modify the statement. To continue, choose Explain Again. Call transaction ST05 and choose Trace list. Select one statement and choose Explain.
46
September 2011
For further analysis of the displayed information, you can choose from various options in the application tool bar. For more information, see EXPLAIN Options [Page 47].
More Information
For additional information about the EXPLAIN function, see SAP Note 400938. For more general information, see the IBM documentation: Administration Guide: Chapter 26, SQL Explain Facility.
DB Catalog
September 2011
47
Additionally, all indexes of the table are displayed with their index columns. Columns Displays selected information from table SYSCAT.COLUMNS for all table columns Indexes Displays information from table SYSCAT.INDEXES for all indexes of the table as well as information from table SYSCAT.COLUMNS for all index columns Update Statistics Updates the catalog statistics for the table If the catalog statistics were updated successfully, the field <stats-time> is displayed in green. Dictionary With this option, you can display the ABAP Dictionary structure (definition) of a table by selecting the table in the access plan. If you do not select a table in the access plan, the ABAP Dictionary structure (definition) of the first dictionary object of the SQL statement is displayed. With this option, you can display the structure of views, even though views never appear in the access plan.
This function is not available for systems monitored using a remote database connection. Test Execution This option is only available, if a: SELECT statement is explained using transaction ST05 Trace list, the parameter values for all parameter markers of the statement are provided and the operation is other than PREPARE SELECT statement without parameter markers is explained When you use the EXPLAIN function, the entered SQL statement is only prepared and the access plan of the optimizer is chosen because of the system catalog statistics. On the basis of this information the optimizer estimates the costs for the execution of this statement. However, the estimated costs may not correspond to the real execution time. Reasons for this might be bad statistics, a bad database layout or problems of the optimizer itself. The Test Execution option measures the real execution time and provides other snapshot data, such as the number of buffer pool accesses and sorts for the selected statement. When the statement is executed, the parameter markers are replaced by the actual parameter values. A dialog box appears where you can change these values to investigate the dependence of the execution time from these values.
48
September 2011
The result of several test executions of the same statement can vary because, for example, the buffer pool may already contain data that is necessary for the execution.
This function is not available for systems monitored using a remote database connection. Tree Info The following additional information is displayed or hidden: num_rows Estimated number of rows (result set) tot_cost Estimated total cost for this statement i/o_cost Estimated I/O cost of the statement This information is also included in the output information when you choose Details. Edit When choosing this option, the system switches to an editor window in which you can modify the selected SQL statement and explain it again. This option is only available when the statement contains a LOCATION comment, for example, when you call EXPLAIN using transaction ST05 Trace list. The location of the statement in the ABAP source code is displayed in a separate window.
Source
This function is only available for the local system or ABAP systems for which an additional RFC destination has been assigned.
The Collect function is no longer available. To collect EXPLAIN data, use the db2support command line tool.
September 2011
49
Activities
You use the index advisor to perform one of the following actions: To evaluate the SQL statements in the dynamic SQL cache by receiving recommendations for potential new indexes that might improve the overall system performance Based on the current content of the dynamic SQL cache, the DB2 Design Advisor determines and recommends new indexes that might improve the overall system performance. For more information, see Retrieving Index Recommendations for the Dynamic SQL Cache [page 51]. To determine and recommend new virtual indexes for a single SQL statement On the basis of a given SQL statement, the DB2 Design Advisor determines and recommends new indexes that might improve the performance of the query. For more information, see Retrieving Index Recommendations for a Single SQL Statement [page 53].
50
September 2011
To create user-defined virtual indexes If you are not satisfied with the recommendations of the DB2 Design Advisor, you create a virtual index specifically tailored to your requirements. For more information, see Defining Virtual User-Defined Indexes [page 56]. To include the indexes in the EXPLAIN function when explaining a query You can check, for example, if the virtually defined indexes would improve the performance of a query. For more information, see Validating Indexes Using the EXPLAIN Function [page 58].
September 2011
51
Column
Description Recommended index that already exists and that the DB2 optimizer would use White Existing index that is, however, not appropriate for the respective SQL statement Red Recommended index that does not yet exist End of the note. o Yes Index exists in the database.
Exists o No Index is a virtual index. Table Name Table on which the index is defined Schema Name of the index schema Specifies a unique rule: o Uniqueness o o NLEAF NLEVELS Unique Duplicates Allowed Primary Key
Number of leaf pages Number of index levels o Yes Index supports reverse scans.
Rev. Scans o No Index does not support reverse scans. Columns INCLUDEs Column Names Number of columns in the key plus the number of included columns if there have been any defined Number of included columns List of column names
52
September 2011
3. To retrieve more information about which SQL statement would benefit from the recommended indexes, select an index from the list. The details are displayed in the following table in the content detail area: Column SQL Statement Frequency Cost Saving Cost Before Cost After Description Name of the SQL statement that is in the dynamic package cache and that would benefit from the index Number of times the statement has been executed since it has entered the dynamic SQL cache Estimated cost savings in percent after the index was created Estimated SQL cost in timerons before the index was created Estimated SQL cost in timerons after the index was created
You can find the complete output of the DB2 design on the Advisor Output tab page. The complete output also includes the estimated space requirements of each recommended index. Note If you do not want to display unused indexes, you can set a filter on the table accordingly. End of the note.
September 2011
53
The results are displayed in the following table: Column Index Name Name of the index o Existing (not-used) Index exists in the system catalog, but for the investigated SQL query it is not identified as usable by the DB2 optimizer. o User-Defined Index has been virtually defined by the user to determine whether such an index could be used to improve the query performance. Those indexes do not exist in the system catalog. o Recommended Index is recommended by the DB2 Design Advisor. Recommended indexes that do not exist are candidates for new indexes to be created. Type Note Indexes are displayed with the following background colors: Green Recommended index that already exists and that the DB2 optimizer would use White Existing index that is, however, not appropriate for the respective SQL statement Red Recommended index that does not yet exist End of the note. o Exists o No Index is a virtual index. Table Name Table on which the index is defined Schema Name of the index schema Yes Index exists in the database. Description
54
September 2011
Description
Duplicates Allowed
Number of leaf pages Number of index levels o Yes Index supports reverse scans.
Rev. Scans o No Index does not support reverse scans. Columns INCLUDEs Column Names Note If you do not want to display indexes that are not used, you can set a filter on the table accordingly. End of the note. Number of columns in the key plus the number of included columns if there have been any defined Number of included columns List of column names
September 2011
55
Prerequisites
You have already retrieved index recommendations for a single SQL statement [page 53].
Procedure
In the Index Advisor screen, choose Add Virtual Index. Depending on your user interface, perform the following steps:
56
September 2011
Result
You can now use the EXPLAIN function to validate existing, recommended, and newly created user-defined indexes as described in the following section.
September 2011
57
Result
A new dialog window or Web browser appears displaying the access plan that the DB2 optimizer considers to be the most efficient one.
Procedure
1. On the Index Advisor screen, choose an index (user-defined or recommended) that does not yet exist. 2. Choose the Create Index in ABAP Dictionary pushbutton (that is located next to the Index Name column). The Create Index in ABAP Dictionary dialog box appears.
58
September 2011
3. Enter a description for the index and choose Create. The index is created in the ABAP Dictionary. After the index has successfully been created, you can decide if you want to schedule a RUNSTATS for the affected table. Note If the index to be created is extending an existing unique index (including primary keys) with one or more INCLUDE columns, you have to replace the existing index with a new index instead of creating a new one. In this case, the Replace Existing Unique Index dialog box appears instead of the Create Index in ABAP Dictionary dialog box. If you replace the existing index, the system replaces the index only on database level. This means that no changes are applied to the ABAP Dictionary. The replacement is automatically scheduled as an SQL script in the DBA Planning Calendar. End of the note.
September 2011
59
4 Performance
4 Performance
The following sections provide information about the main task areas that are available under Performance on the Database tab page of the DBA Cockpit: Performance Warehouse Time Spent Analysis Workload Statistics Top SQL Statement Analysis Snapshots Critical Activities Utilities History
Prerequisites
An SAP Solution Manager system with Solution Manager Diagnostics (SMD) enabled is required.
Integration
The Performance Warehouse is part of the DBA Cockpit.
Features
In the Performance Warehouse, all relevant performance indicators that are collected by the DBA Cockpit are stored in an SAP NetWeaver Business Warehouse system. This SAP NetWeaver BW system is used by the Solution Manager Diagnostics (SMD) back-end of an SAP Solution Manager system. SMD already uses this SAP NetWeaver BW to store workload data of SAP applications. To configure the extraction of data into the SMD BI, you use the SMD Setup Wizard. Based on this architecture, the DBA Cockpit uses SAP NetWeaver BW technology to provide reports for performance analysis, which you can customize according to your needs. All collected data has a time dimension, so you can analyze the database performance for any point in time or over a specified time frame.
60
September 2011
4 Performance
Almost all reports are displayed as a chart to visualize the key performance indicators (KPIs). In addition, there is also a detailed table view. To navigate within these reports, you can use the SAP NetWeaver BW drilldown feature. Violations to performance thresholds are highlighted based on predefined BW exceptions to make you immediately aware of performance issues. By default, the Performance Warehouse is delivered with predefined content that you can use to create your own reports according to your needs.
More Information
Performance Warehouse: Reporting Performance Warehouse: Configuration
Displaying a Report
To display a report, choose the appropriate view pushbutton on the respective tab page. Note The availability of the tab pages and of the pushbuttons on each tab page can vary depending on the selected system. Some reports are only available if special database features are enabled. End of the note.
September 2011
61
4 Performance
The reports consist of two sections: In the upper section, a chart is displayed to visualize the key performance indicators. The chart provides a subset of the key columns from the detail table view. Note The chart display is optional and not available for all views. End of the note. In the lower section, a detailed table view is available. You can drill down your reports by either using the context menu of a column header in the Detail: <Category View> screen area or by specifying the respective value using the pushbuttons in the Detail: Navigation screen area. Here, you can also add and remove columns or key figures, or you can set filters on columns. In addition, there are predefined exceptions (for example, Chart: Exceptions or Details: Exceptions) for almost all reports on key performance indicators. The used thresholds are based on Early Watch Alerts and each violation to these thresholds is displayed in red. Note If you want to reset a report to its initial state, choose Reset Report in the central system area. End of the note.
62
September 2011
4 Performance
Configuration
Here, you can view or modify the configuration parameters of the performance warehouse for the monitored system. To modify some of these parameters, use the Edit, Save, and Cancel pushbuttons in the toolbar. Depending on your database platform, the displayed selection of values can vary. The following parameters are displayed for all database platforms: Parameter BI Server Reporting Time Zone Note The Default checkbox is selected if the default value for your complete landscape is the same as the one specified for your system. End of the note. Description BI server where the database performance data is located The performance data time-stamps are converted to one global time zone for all reports in SMD BI.
Web Reports
Here, you can configure the display on the Reporting screen. That is, you can view and modify the integrated BI BEx Web templates in the tree table. To modify some of these parameters, use the Edit, Add, and Delete pushbuttons in the toolbar. The main report categories appear and for each report category, you can view or modify the views by expanding the appropriate report category. These views appear as pushbuttons on the respective category tab page on the Reporting screen. To change the sequence within a category, use the Up or Down pushbuttons. To display details about a view, simply select it in the table. The following parameters are displayed in the Details for Web Reports area below the table view: Parameter Description Specifies the name of the report Report This text appears on the view pushbutton on a category tab page. Detailed description for the report Description This text appears as a tooltip for the pushbutton of the key indicator on a category tab page. Specifies the report category Category Each category is represented on a separate tab page. Web Report (Default) Technical name of the BI BEx Web templates
September 2011
63
4 Performance
Parameter
Description
Web Report (Day) Technical name of the BI BEx Web templates for granularity Day Web Report (Month) Technical name of the BI BEx Web templates for granularity Month Specifies the data provider of the BI BEx Web templates with a time dimension The drilldown of the time dimension is changed according to the selected granularity. If selected, the report is available for performance analysis. If selected, the report is executed as soon as the tab page is selected. Name of database platform Minimum database release for this report Maximum database release for this report
Depending on your database platform, there might be more checkboxes available for database-specific features. If you select these checkboxes, the report is only displayed if the specific database features have been set up in the monitored database system. End of the note.
Report Categories
Here, you can view and modify the categories for BI BEx Web templates of the reports that are displayed the Reporting screen. To modify some of these parameters, use the Edit, Add, and Delete pushbuttons in the toolbar. To change the sequence of the categories on the Reporting screen, use the Up or Down pushbuttons. The following parameters are displayed: Parameter Category Description Name of the category Detailed description for the category Description
64
September 2011
4 Performance
September 2011
65
4 Performance
Description
If you choose this option, real-time monitoring data is collected. This means that specific workload situations can be analyzed in real-time. To set a time interval for the data collection and the charts to be refreshed, choose an interval from the Refresh Interval dropdown list. To start the ad-hoc data collection, choose the Resume pushbutton. If you want to change the drill-down or filter criteria, you must first stop the current Ad hoc data collection by choosing the Pause pushbutton. Time Frame You specify the time interval and time frame for which you want data to be analyzed. Aggregate If you choose the Aggregate option from the dropdown list, a pie chart is displayed according to your selected filter and drill-down criteria. Type of Display Timeline If you choose the Timeline option from the dropdown list, the monitoring data is displayed in a histogram according to your filter and drill-down criteria. You can condense the granularity of the time interval from As Collected to Hour, Day, Week, or Month. Note You can only filter by members if you are using a distributed database system. Members End of the note. From the dropdown list, you can choose the database member on which you want to apply the specific workload filter. You can choose all or single members. If you want to combine multiple members into a group (for example, by physical host), you can use the Advanced Selection option. From the dropdown list, you can choose the service classes that have been set up on the system. You can choose all or single service classes. If you want to combine multiple service classes into a group, choose Advanced Selection from the dropdown list.
Service Class(es)
66
September 2011
4 Performance
From the dropdown list, you can choose between the following time-related WLM metrics: All All known time-related metrics are included. All (Wait Times) All wait time metrics are selected (W) All (Processing Times) All processing times without wait times are selected (P). Single metrics, for example, (P) Commit Processing Time or (W) Buffer Pool Read Time For more information, see Monitoring metrics for DB2 workload manager in the IBM Information Center. Advanced Selection When you choose this option, all metrics including their parent metrics are displayed. You can now select those metrics that are relevant for your analysis. For example, you select all REORG and RUNSTATS processing times to drill down to specific members or member groups. If you choose the Choose Level of Detail link, you can specify one of the following options in the Metrics Hierarchy dialog box: Total Request Time Only Aggregates all metrics Wait vs. Processing vs. Unknown Times Aggregates metric groups Detailed Wait and Processing Time Includes only detailed wait and processing times Full Details Includes all time metrics (default) Note The metrics are mutually exclusive and are independent of each other. Since all metric metadata is retrieved dynamically from the database, the availability of specific metrics as filter criteria depends on the database release. End of the note.
Metric(s)
September 2011
67
4 Performance
You can drill down into the display by choosing one of the following options from the dropdown list: Members All results are distributed across the different members based on the specified filter criteria. Note You can only drill down by members if you are using a distributed database system. End of the note. Metrics All results are displayed according to the different time-based monitoring metric occurrences and their relative values as a percentage of the total request time. Service Classes Based on the specified filter criteria, all results are distributed across the service classes according to their occurrence and their relative percentage to the total request time. To avoid the chart becoming overloaded and unclear, the Reduce Display to Most Significant checkbox is by default selected. All metrics, members, and service classes with insignificantly small values are cumulated to give the value of Other Known Times. Note Make sure that drill-down and filter criteria are not the same, that is, do not use, for example, non-aggregated single metrics combined with the option to drill down by metrics. End of the note.
Drill-Down
68
September 2011
4 Performance
Timeline histogram The timeline-based chart lets you analyze changes in workload performance for a given time frame. This chart type displays the cumulated occurrences of selected time-based metrics, members, or service classes into a histogram that consists of bars that are aligned to the timeline. The histogram bars are divided according to the time granularity you have specified in the Selection area.
September 2011
69
4 Performance
Example 2
To identify the reason for a general system slowdown, perform the following steps: 1. As Type of Display, choose Timeline, and as Drill-Down option, choose Metrics. 2. In the Members and Service Class(es) fields, choose All and apply your selection. 3. Identify the most significant time metrics in the displayed chart. 4. In the detailed Summary area of those metrics, check for outlier members or service classes to restrict the scope to a smaller group of members or service classes. 5. For a more detailed analysis of the system slowdown, keep your selection of members and service classes and navigate into other areas of the DBA Cockpit such as the following: o Performance Snapshots SQL Cache
Search for statements with high values for the identified metrics. o Diagnostics Lock-Wait Events
Analyze statistical metrics for the identified members or service classes. o Configuration Workload Management Workloads and Service Classes
More Information
Monitoring metrics for DB2 workload manager in the IBM Information Center at: https://publib.boulder.ibm.com/infocenter/db2luw/v9r7/index.jsp?topic=/com.ibm.db2.luw.admi n.wlm.doc/doc/c0055265.html
70
September 2011
4 Performance
Activity Lifetimes
The histograms that you can choose from the Histogram Type dropdown list give you a better understanding of your database workload. The following histogram types are available: Activity Execution Time Histogram Displays the history of the histogram for the execution times of all activities that were executed in the selected service classes and members for the selected time frame The execution time corresponds to the time an activity has spent executing on the coordinator partition. Execution time does not include the time spent during initialization, queuing, or between cursor operations.
September 2011
71
4 Performance
Activity Lifetime Histogram Displays the history of the histogram for the lifetimes of all activities that were executed in the selected service classes and members for the selected time frame The lifetime corresponds to the total elapse time of an activity, as measured on the coordinator partition. Lifetime does include the time spent during initialization, queuing, or between cursor operations. Activity Queue Time Histogram Displays the history of the histogram for the queue times of all activities that were executed in the selected service classes and members for the selected time frame The queue time corresponds to the time an activity has spend in a workload management (WLM) queue. The average execution, life, or queue time for all activities is displayed in a dotted line. The total number of activities that have completed, were rejected, or aborted by WLM in the selected time frame is displayed in an additional chart below the histogram chart. For more information about the histogram types and all other monitoring counters provided here, see the IBM DB2 Information Center for your database release at: http://www-01.ibm.com/support/docview.wss?rs=71&uid=swg27009474
Statistics
The Statistics tab page provides you with a range of high-watermark and average metrics that are separated by service class, member, and time. By default, the following information is displayed: Column Time from Time to Member Service Superclass Service Subclass Coord. Act. Lifetime Top Coord. Act. Lifetime Avg. UOW Total Time Top Description Begin of statistics interval timestamp End of statistics interval timestamp Member that has captured the statistics Name of the service superclass for which the statistics were captured Name of the service subclass for which the statistics were captured High-water mark in milliseconds for coordinator activity lifetime
Arithmetic mean of lifetime in milliseconds for coordinator activities High-water mark in milliseconds for unit of work lifetime
Act. Temp. Tablespace High-water mark in KB for the temporary tablespace usage of a Top single DML activity
72
September 2011
4 Performance
Description High-water mark in KB for the aggregate temporary tablespace usage across all DML activities High-water mark for processor time used by activities at all nesting levels, in milliseconds High-water mark for the number of rows read by activities on all nesting levels High-water mark for the number of rows returned by activities on all nesting levels
For more Information about all available statistics, see the IBM DB2 Information Center for your database release at: http://www-01.ibm.com/support/docview.wss?rs=71&uid=swg27009474
September 2011
73
4 Performance
In the Selection area, you specify the appropriate time frame. The result is an aggregated view of the metrics for the SQL statements captured in the specified time frame. To limit the result set, you can specify ranking or filter criteria in the Selection area by either choosing the Top SQL Statements by or Custom radio button. If you choose the Top SQL Statements by option, you have to choose a ranking criterion from the relevant dropdown list. If you choose the Custom radio button, you can specify filter criteria in the relevant fields. In the Selection area, you can also define the maximum number of returned SQL statements. The default limit is 100 SQL statements. By default, static packages are not displayed as the SAP system uses dynamic SQL only. If you want to analyze static packages, you have to select the Display Static Packages checkbox. If you choose the Apply Selection pushbutton, the result set is determined according to your selection criteria. The following information is displayed: Column Statement Text Text of the SQL statement The first metric column is used for ranking as chosen in the Selection area from the Top SQL Statements by drop-down list. By default, Coord. Statement Exec. Time is used. No. of Coord. Executions Number of executions by coordinator agent Description
<ranking metric>
Avg. Coord. Average execution time in milliseconds for statement by coordinator agent Stmt. Exec. Time Total CPU Time Total CPU time in milliseconds Ratio of rows read from the base table compared to rows processed, which can be either rows returned to the application (SELECT statements) or rows written (UPDATE, INSERT, DELETE statements) A value of 1 indicates optimal access to the requested data. High values indicate statements with inefficient access. BP Gets / Rows Processed Average number of pages read from the buffer pool per rows processed, which can be either rows returned to the application (SELECT statements) or rows written (UPDATE, INSERT, DELETE statements)
74
September 2011
4 Performance
SQL Statement
On the SQL Statement tab page, the complete SQL statement text is displayed. To show the current access plan of the statement, you can choose the EXPLAIN pushbutton. For more information, see The EXPLAIN Function [page 43]. To display the ABAP source code from where the SQL statement was executed, you can choose the Show Source pushbutton. Note The Show Source pushbutton is only available if the following applies: The SAP system on which the currently monitored database is running is an ABAP system. The SAP system has the minimum Support Package level as described in SAP Note 1493490. End of the note. In contrast to the analysis of the SQL cache, the compilation environment is not available at this level because you see an aggregation of metrics over a time frame that could have multiple compilations of the SQL statement. For more information about the SQL cache, see Snapshots: SQL Cache (DB2 V9.7 or higher only) [page 102].
Metrics Summary
The Metrics Summary tab page provides the chart view Time Spent in SQL Statement, which is an aggregated view of the time frame defined in the Selection area. In addition, detailed information about the following is displayed: Statement metadata Miscellaneous metrics Sorts Locking Logger Buffer pool Direct I/O
September 2011
75
4 Performance
Metrics Details
The Metrics Details tab page provides the selection of specific time slices. The detailed information displayed is a view of the specific time slice: Statement metadata Miscellaneous metrics Sorts Locking Logger Buffer pool Direct I/O
76
September 2011
4 Performance
SQL Statement
On the SQL Statement tab page, the SQL statement text and the compilation environment are displayed. To show the access plan of the statement, you can choose the Historic EXPLAIN pushbutton. Historic here means that it shows the access plan captured when the access plan was generated. For more information, see The EXPLAIN Function. Note The Historic EXPLAIN is only available if the corresponding option Collect Section EXPLAIN is enabled for the DCF data collector Package Cache. In the SAP Default or SAP Detailed templates this option is not enabled. End of the note. The Compilation Environment area displays information, for example, about the following: Isolation Query optimization SQL rules Refresh age Resolution time stamp
Metrics Summary
The Metrics Summary tab page provides the chart view Time Spent in SQL Statement. It is an aggregated view over the timeframe defined in the selection area and the access plan version selected. In addition, detailed information about the following is displayed: Statement metadata Miscellaneous metrics Sorts Locking Logger Buffer pool Direct I/O
September 2011
77
4 Performance
Metrics Details
The Metrics Details tab page provides the selection of specific time slices for the access plan version. The detailed information displayed is a view of the specific time slice: Statement metadata Miscellaneous metrics Sorts Locking Logger Buffer pool Direct I/O
Features
In the Snapshots task area, monitoring data in both unfiltered and raw format is provided by the various snapshots. Snapshot data is provided on different levels (for example, database, buffer pools, or SQL statements) and is collected by the database beginning from startup of the database instance until the current time. That is, data is accumulated over a long time
78
September 2011
4 Performance
period and, therefore, cannot reflect short-term decreases in performance or increasing or changing workloads over a certain time period. Furthermore, for the interpretation of snapshot data, it is necessary to refer to related configuration parameters. Configuration parameters, however, do not relate to a certain time frame either but only describe the current configuration of the database. Especially if your database is enabled for self-tuning memory management (STMM), there is no way of correlating bad performance data with current storage parameters. To overcome these problems, you can observe monitoring data over a certain time period by doing one of the following: You can explicitly set a starting point for measurements using the Reset and Since Reset pushbuttons that are available for most snapshot options. As a result, a set of monitoring data is displayed that is related to the time frame defined by the reset operation. In parallel, you can check related configuration parameters for changes by choosing Configuration Configuration History on the Database tab page of the DBA Cockpit. If the relevant snapshot option supports the evaluation of snapshot data that is periodically collected by the data collection framework (DCF), you can specify any time interval for analysis as long as the relevant data is kept in the DCF history. Since the displayed configuration data is also retrieved from the DCF, there is a consistent, time-related view of both the snapshot and the configuration. This applies to the following snapshot options: o o o o o o o Database Buffer Pools Tablespaces Containers Tables SQL Cache System
September 2011
79
4 Performance
Features
The Database screen provides the following: A selection area where you specify filter criteria for the monitoring data (only available in a distributed database) An overview table displaying the retrieved monitoring data with the most important key figures that have been aggregated over the selected time period. A detail view of selected monitoring data of the result list An option to break down all history data that was collected for the selected monitoring data during the specified time frame.
80
September 2011
4 Performance
Process
To analyze database snapshot data, you proceed as follows: 1. You identify the time period where you experienced performance problems, such as, long-running background jobs, threshold violations detected by workload management, and bad user response time. 2. In the Selection area, you specify the appropriate time frame. Note If you are using a distributed database system and you want to retrieve a general snapshot, you set the member selection to All and you do not drill down by members. To retrieve a more detailed snapshot in a distributed database system, you can further drill down the selection by members. If a single member or a subset of members shows deviations compared to other members, you can also restrict the member selection. End of the note. 3. To refresh the monitoring data, you choose the Apply Selection pushbutton. The following information is displayed for each partition: o o o o o o o Number and total size of buffer pools Number of data and index logical reads Number of data and index physical reads Average physical read and write time that is required to read from and write data into the buffer pool Executed SQL statements Package cache size Package cache quality
4. To display detailed information, you select a line in the overview table. For more information, see Displaying Detailed Database Snapshot Data [page 82]. 5. To display detailed history data, you select a line in the overview table and choose the History pushbutton. The data is displayed in the History Details content detail area. 6. To isolate the time frame during which the performance problem occurred, you use the data in the overview table in the History Details content area.
September 2011
81
4 Performance
2. In the content detail area of the Database screen, choose the appropriate tab page to display an overview of the following critical database performance indicators: o o o o o o Buffer Pool [page 295] Cache [page 297] Asynchronous I/O [page 298] Direct I/O [page 299] Real-Time Statistics [page 300] Locks and Deadlocks [page 300]
82
September 2011
4 Performance
o o o o
Logging [page 302] Calls [page 304] Sorts [page 306] XML Storage [page 307]
The data displayed on these tab pages is described in more detail in the following sections.
September 2011
83
4 Performance
Description Percentage of index physical read accesses for the component (and partition if you are using a multi partition database)
Features
The enhanced Buffer Pool screen provides the following: A selection area where you specify filter criteria for the monitoring data (only available in a distributed database) An overview table displaying the retrieved monitoring data with the most important key figures that have been aggregated over the selected time period. A detail view of selected monitoring data of the result list An option to break down all history data that was collected for the selected monitoring data during the specified time frame.
84
September 2011
4 Performance
Process
To analyze buffer pool snapshot data, you proceed as follows: 1. You identify the time period where you experienced performance problems, for example, long running background jobs, threshold violations detected by workload management, and bad user response time 2. In the Selection area, you specify the appropriate time frame. Note If you are using a distributed database system and you want to retrieve a general snapshot, you set the member selection to All and you do not drill down by members. To retrieve a more detailed snapshot in a distributed database system, you can further drill down the selection by members. If a single member or a subset of members shows deviations compared to other members, you can also restrict the member selection. End of the note. 3. To refresh buffer pool snapshot data, you choose the Apply Selection pushbutton. An overview table provides information about important key figures of the buffer pool activity of your database, for example: o o o o o o o Buffer pool size Members Percentage at which the data is read from the buffer pool, rather than directly from the hard disk Whether the buffer pool is enabled for STMM Number of data and index logical reads Number of data and index physical reads Number of logical and physical read requests that required I/O to get data pages into the temporary tablespace
4. To display detailed information about a buffer pool, you select a line in the overview table. For more information, see Displaying Detailed Buffer Pool Snapshot Data [page 86]. 5. To display detailed history data, you select a line in the overview table and choose the History pushbutton. 6. To isolate the time frame during which the performance problem occurred, you use the data in the overview table in the Buffer Pool History content area.
September 2011
85
4 Performance
2. In the content detail area of the Buffer Pools screen, choose the appropriate tab page to display an overview of the following critical database performance indicators: o o o o Buffer Pool [page 308] Asynchronous I/O [page 310] Direct I/O [page 311] XML Storage [page 312]
The data displayed on these tab pages is described in more detail in the following sections.
86
September 2011
4 Performance
Features
The Tablespaces screen provides the following: A selection area where you specify filter criteria for the monitoring data (only available in a distributed database) An overview table displaying the retrieved monitoring data with the most important key figures that have been aggregated over the selected time period. A detail view of selected monitoring data of the result list An option to break down all history data that was collected for the selected monitoring data during the specified time frame.
Process Flow
To analyze tablespace snapshot data, you proceed as follows: 1. You identify the time period where you experienced performance problems, for example, long running batch jobs, threshold violations detected by workload management, and bad user response time 2. In the Selection area, you specify the appropriate time frame.
September 2011
87
4 Performance
Note If you are using a distributed database system and you want to retrieve a general snapshot, you set the member selection to All and you do not drill down by members. To retrieve a more detailed snapshot in a distributed database system, you can further drill down the selection by members. If a single member or a subset of members shows deviations compared to other members, you can also restrict the member selection. End of the note. 3. To refresh tablespace snapshot data, you choose the Apply Selection pushbutton. An overview table provides information about important key figures of the tablespace activity, for example: o o o o o o o Tablespace size Members Buffer quality Average physical read and write times The number of data and index logical reads The number of data and index physical reads Name of the buffer pool the tablespace is assigned to
4. To display detailed data about a tablespace, you select a line in the overview table. For more information, see Displaying Detailed Tablespace Snapshot Data [page 89]. 5. To display detailed history data, you select a line in the overview table and choose the History pushbutton. 6. To isolate the time frame during which the performance problem occurred, you use the data in the overview table in the Tablespace History content area.
88
September 2011
4 Performance
2. In the content detail area of the Tablespaces screen, choose the appropriate tab page to get an overview of the following critical database performance indicators: o o o o Buffer Pool [page 313] Asynchronous I/O [page 315] Direct I/O [page 316] XML Storage [page 316]
Buffer Pool and Asynchronous I/O provide information about buffer pool access. Direct I/O and XML Storage provide information about direct accesses, that is, I/O activity that does not use the buffer pool (for example, access to LONG VARCHAR columns or backup).
September 2011
89
4 Performance
You can use the Containers snapshot screen to gain an overview of the containers in your database. To access the container snapshot, call the DBA Cockpit and choose Performance Snapshots Containers on the Database tab page of the DBA Cockpit. The Containers screen provides the following: A Selection area where you specify filter criteria for the collected data In the Sort Tables by field, you can choose if the containers are sorted by direct reads or by accessibility. An overview table displaying the retrieved history data according to your selection criteria In the overview table on the Containers screen, the following information is displayed: Column Tablespace Name Container Name Name of the tablespace Name of the container Indicates whether the container is accessible Possible values are: Accessibility 1 : Container is accessible. 0 : Container is not accessible. Stripe Set Direct Writes Stripe set of the container Number of write operations that do not use the buffer pool for the container Number of read operations that do not use the buffer pool for the container Elapsed time for direct reads in milliseconds Description
Direct Reads Direct Read Time (ms) Direct Write Time (ms) Est. Time per Write (ms) Pages Read
Elapsed time for direct writes in milliseconds Estimated time per write operation (direct writes and direct write time) in milliseconds Number of pages read from the container
90
September 2011
4 Performance
Depending on the setup of your DBA Cockpit, one of the following snapshots is displayed on the Tables screen: Enhanced table snapshot with support of the data collection framework (DCF) Table snapshot without DCF support For more information about the DCF, see Enabling the Database for the Data Collection Framework [page 32]. The Tables screen provides information that you can use to analyze tables under the following aspects: 1. Which tables are accessed the most and require extra tuning? 2. Which tables have frequent update operations and are, therefore, potential candidates for new statistics? 3. Which tables are candidates for reorganization due to space fragmentation?
Features
The Tables screen provides the following areas: A selection area where you specify filter criteria for the monitoring data (only available in a distributed database system) A result list of the retrieved monitoring data providing the most important key figures that have been aggregated over the selected time period An option to break down all history data that was collected for the selected monitoring data during the specified time frame
September 2011
91
4 Performance
Process
To analyze table snapshot data, you proceed as follows: 1. You identify the time period where you experienced performance problems for example, long running background jobs, threshold violations detected by workload management, and bad user response time 2. In the Selection area, you specify the appropriate time frame. Note If you are using a distributed database system and you want to retrieve a general snapshot, you set the member selection to All and you do not drill down by members. To retrieve a more detailed snapshot in a distributed database system, you can further drill down the selection by members. If a single member or a subset of members shows deviations compared to other members, you can also restrict the member selection. End of the note. 3. You restrict the result set in the overview table, for example, by limiting the maximum number of rows, by using wildcards, or by specifying lower limits for I/O counters. Example By using the appropriate wildcard in the Table Name input field, you can view special BW tables with a specific naming convention. End of the example. 4. To refresh table snapshot data, you choose the Apply Selection pushbutton. 5. To analyze a single table in more detail, you select a line in the overview table and choose the Detailed Table Analysis pushbutton. You are redirected to the Single Table Analysis screen where detailed table analysis data is displayed. 6. To display detailed history data, you select a line in the overview table and choose the History pushbutton. The data is displayed in the History Details content area. 7. To isolate the time frame during which the performance problem occurred, you use the data in the overview table in the History Details content detail area.
92
September 2011
4 Performance
Name of the table Number of partition (only displayed in a multi-partition database) File ID (FID) of the table Type of table for which information is displayed, for example, user, system, or temp Number of rows changed (inserted, deleted, or updated) in the table
Rows Read Number of rows read from the table Number of accesses (reads and writes) to overflowed rows of the table Overflowed rows indicate that data fragmentation has occurred. If this number is high, you might be able to improve table performance by reorganizing the table using the REORG utility, which cleans up this fragmentation. Overflow Access Note Pay particular attention to this column. If the value in this column is very high, consider reorganizing the table. End of the note. Number of page reorganizations executed for the table Page REORGs Too many page reorganizations can result in less than optimal insert performance. You can use the REORG TABLE utility to reorganize a table and eliminate fragmentation.
Note If you select a line, you are redirected to the Single Table Analysis screen where detailed table analysis data is displayed. End of the note.
More Information
Analyzing and Maintaining a Single Table [page 154]
September 2011
93
4 Performance
Process
To analyze application snapshot data, you proceed as follows: 1. If the currently monitored database comprises more than one member, you first specify in the Selection area whether you want to retrieve information from applications that are connected to any member or only to certain members by setting the appropriate filter criteria. Note At the beginning of a unit of work (UOW), each application is assigned to a workload and therefore, also to a service class. This assignment is reevaluated at the beginning of the next UOW. If you want to monitor applications that are assigned to a specific service class, for example, the ABAP dialog service class, set a filter accordingly in the Selection area. End of the note. In addition, you specify if you want to display information about DB2 autonomic processes (for example, STMM or Event Monitor) by selecting the Show DB2 Autonomics and Utilities checkbox.
94
September 2011
4 Performance
To refresh the monitoring data, choose the Apply Selection pushbutton. The following information is displayed: Column Description A system-wide unique ID for the application in a distributed database system This ID is the same on every member where the application can span subagents. ID of the coordinating member of the application
Application Handle
Coord. Member
Application State State of the application Application Name UOW Runtime Entity Client PID Client Workstation Name of application running at the client as known to the database manager or DB2 Connect Total runtime of the current unit of work (UOW) in milliseconds Type of the entity (db2agent, db2fmp, or name of the system entity) Process ID of the client application that made the connection to the database Client workstation as specified in the TPMON special register Client user ID as specified in the TPMON special register. Client User ID In an SAP ABAP system, the client user ID corresponds to the SAP user name. Client application as specified in the TPMON special register In an SAP ABAP system, the client application is specified as follows: o Client Application o o ptype BTC for SAP batch work processes ptype SPO for SAP spool work processes ptype UPD or ptype UPD2 for SAP update work processes the SAP transaction code for SAP dialog work processes
Name of the service superclass to which the coordinating agent is currently assigned Name of the service subclass to which the coordinating agent is currently assigned
September 2011
95
4 Performance
2. To set a filter on the retrieved snapshot data, you specify the following in the Summary area: Note The numbers displayed behind the selectable characteristics indicate the number of applications, agents, or activities that are currently in the respective state. End of the note. Characteristic State of an application The following states are possible: o Active The application is currently executing activities on the database. o Application States o Inactive (having an uncommitted activity) The application is currently not executing activities on the database. However, the application has done so in the past and these past activities have not yet been committed. Inactive (no uncommitted activity) The application is currently not executing any activities on the database and has committed all activities that were executed in the past. o Inactive (no agent assigned) No coordinator is assigned to the application. Activity types that are currently being executed on the database The following values are possible: o o o Activity Types o o o Note Only activity types that are currently processed are displayed. End of the note. LOAD READ DML WRITE DML DDL CALL OTHER Description
96
September 2011
4 Performance
Characteristic
Description States of the activities that are currently being executed on the database Possible values include, for example, the following states: o o o Cancel Pending Executing Idle Initializing Queued Terminating
Activity States
o o o Note
Only activity types that are currently processed are displayed. End of the note. Events that have recently been processed by agents that are assigned to any of the application handles Possible values include, for example, the following activity states: o o Events Recently Processed o o o Note Only activity types that are currently processed are displayed. End of the note. 3. To trace, capture, or cancel an activity, you select an application handle in the overview table of the Application screen and choose the respective pushbutton as described in Tracing, Capturing or Canceling an Activity later in this section. 4. To display a detailed view of a specific application, you choose an application handle from the result list. For more information, see Displaying Details about Application Handles later in this section. Process Request Process Routine Wait Request Acquire Lock Wait WLM Queue
September 2011
97
4 Performance
End of the note. To review the status of all assigned agents at a certain point in time, you have to pause the trace. After selecting the respecting point in time in the Unit of Work Progress table, details about the assigned agents of that point in time is again displayed in Assigned Agents table. The trace is automatically paused as soon as the application disconnects from the database system or as soon as this window is closed. Capture detailed information about the currently executed activity of an application as follows: 1. Select the application and choose the Capture Activity pushbutton. 2. In the DBA Cockpit, choose Violations . Performance Critical Activities Threshold
On the Threshold Violations screen, activities that have been captured manually are displayed with the value MANUAL in the Predicate column.
98
September 2011
4 Performance
Note To find manually captured activities more easily, use the filter function of the list viewer. End of the note. Cancel an activity by selecting the application and choosing the Cancel Activity pushbutton. If the activity is successfully canceled, the SQL error SQL4725N with status SQLSTATE 57014 is returned to the canceled application.
Unit of Work
On the Unit of Work tab page, you can find details about the unit of work (UOW) that the application is currently executing on the database system. The pie chart indicates where this UOW has spent most of its time since it was started by the respective application. Note If the selected UOW has just started and therefore no metrics have yet been rolled up into the UOW, the following message is displayed instead of the pie chart: No data is available End of the note. You can find more details in the chart area below, for example, detailed aggregate metrics about all activities that were or are just running as part of this unit of work as well as details on the UOW itself and its WLM context.
Current Activities
The Current Activities tab page displays the activities that are currently processed as part of the current UOW. In most cases, only a single activity (identified by an activity ID) is displayed. Only in certain cases for example, if the application has opened a cursor that uses the WITH_HOLD attribute might more than one activity be displayed. If you are using a distributed database or if the activity uses intra-parallelism, you can review the status of each activity on individual members by selecting the relevant row in the table. The details shown below also include the SQL statement text of the activity.
September 2011
99
4 Performance
Assigned Agents
The Assigned Agents tab page displays details about all database agents that are currently assigned to the selected application. In a distributed database or if the activity uses intraparallelism, you can check the status of each individual agent that is currently assigned to the selected application.
100
September 2011
4 Performance
On the Threshold Violations screen, activities that have been captured manually are displayed with the value MANUAL in the Predicate column. Note To find manually captured activities more easily, use the filter function of the list viewer. End of the note. Cancel an activity by selecting it and choosing the Cancel pushbutton. If the activity is successfully canceled, the SQL error SQL4725N with status SQLSTATE 57014 is returned to the canceled application.
September 2011
101
4 Performance
Agents Memory Note This tab page is only available if you are using the Web browser-based user interface. End of the note. Buffer Pool Direct I/O XML Storage Locks and Deadlocks Calls Sorts Cache Unit of Work Statement Statement Text SQL Workspace
102
September 2011
4 Performance
In the Selection area, you can also define the maximum number of returned SQL statements. The default limit is 100 SQL statements. By default, static packages are not displayed as the SAP system uses dynamic SQL only. If you want to analyze static packages, you have to select the Display Static Packages checkbox. If you choose the Apply Selection pushbutton, the result set is determined according to your selection criteria. The following information is displayed: Column Statement Text Text of the SQL statement Description
The first metric column is used for ranking as chosen in the Selection area <ranking metric> from the Top SQL Statements by dropdown list. By default, Coord. Statement Exec. Time is used. No. of Coord. Executions Number of executions by coordinator agent
Avg. Coord. Average execution time in milliseconds for statement by coordinator agent Stmt. Exec. Time Total CPU Time Total CPU time in milliseconds Ratio of rows read from the base table compared to rows processed, which can be either rows returned to the application (SELECT statements) or rows written (UPDATE, INSERT, DELETE statements) A value of 1 indicates optimal access to the requested data. High values indicate statements with inefficient access. BP Gets / Rows Processed Average number of pages read from the buffer pool per rows processed, which can be either rows returned to the application (SELECT statements) or rows written (UPDATE, INSERT, DELETE statements)
SQL Statement
On the SQL Statement tab page, the SQL statement text and the compilation environment are displayed. To show the access plan of the statement, you can choose the EXPLAIN pushbutton. For more information, see The EXPLAIN Function [page 43]. To display the ABAP source code from where the SQL statement was executed, you can choose the Show Source pushbutton.
September 2011
103
4 Performance
Note The Show Source pushbutton is only available if the following applies: The SAP system on which the currently monitored database is running is an ABAP system. The SAP system has the minimum Support Package level as described in SAP Note 1493490. End of the note. The Compilation Environment area displays information, for example, about the following: Isolation Query optimization SQL rules Refresh age Resolution time stamp
Statement Metrics
The Statement Metrics tab page provides the chart view Time Spent in SQL Statement. In addition, detailed information about the following is displayed: Statement metadata Miscellaneous metrics Sorts Locking Logger Buffer pool Direct I/O
104
September 2011
4 Performance
Average execution time in milliseconds for a statement Number of rows read for a statement Number of rows written by a statement Search using either the wild card "*" or using a text string, for example, INSERT, to limit the number of statements displayed Ratio of rows read from the base table compared to rows processed, which can be either rows returned to the application (SELECT statements) or rows written (UPDATE, INSERT, DELETE statements)
A value of 1 indicates an optimal access to the requested data. High Rows Read / values indicate statements with an inefficient access. Rows Processed Note This metric is only evaluated if your database is DB2 V9.5 Fix Pack 1 or higher.
September 2011
105
4 Performance
Description
If you choose the Apply Selection pushbutton, the result set is determined by filtering the snapshot results according to your selection criteria and the following information is displayed in the Summary area: Column Execution Time Rows Read Rows Written Note The data displayed in the Summary area refers to the entire SQL cache and not only to the currently selected result set. End of the note. The result set is displayed in a table as follows: Column SQL Text Description Text of a dynamic SQL statement that was in the SQL cache at the time of the snapshot Number of times a statement was executed This value helps you to identify which statements are executed very often. A high number of executions does not necessarily mean that a statement is using an excessive amount of resources. You should also check the number of rows read and rows written. If you find relatively high values here, choose EXPLAIN to check whether indexes are not being used efficiently or whether indexes are missing. Total execution time in milliseconds for a statement You can use this value together with Executions to identify the statements that would benefit from further analysis. Total Execution Time (milliseconds) divided by Total Cache Sum Execution Time (milliseconds) Description Total execution time in milliseconds for an SQL statement Total number of rows read Total number of rows written
Executions
Total Execution Time (ms) Total Execution Time ( %) Average Execution Time (ms)
106
September 2011
4 Performance
Buffer Quality Buffer quality for this statement in percent (%) Ratio of rows read from the base table compared to rows processed, which can be either rows returned to the application (SELECT statements) or rows written (UPDATE, INSERT, DELETE statements) Rows Read / Rows Processed A value of 1 indicates an optimal access to the requested data. High values indicate statements with an inefficient access. Note This metric is only available if your database is DB2 V9.5 for Linux, UNIX, and Windows Fix Pack 1 or higher. End of the note. Average number of pages read from the buffer pool per rows processed, which can be either rows returned to the application (SELECT statements) or rows written (UPDATE, INSERT, DELETE statements) BP Gets / Rows Processed Note This metric is only available if your database is DB2 V9.5 for Linux, UNIX, and Windows Fix Pack 1 or higher. End of the note. BP Gets / Executions Average number of pages read from the buffer pool per execution of the statement Number of rows read Rows Read You can use this value to identify statements that would benefit from additional indexes. Use EXPLAIN to analyze the statement. The given value does not necessarily correspond to the number of rows of the result set of the SQL statement. The Rows Read value shows the number of rows that needs to be read to obtain the result set. Number of rows that were changed (inserted, deleted, or modified) in a table Rows Written High values might indicate that you should update statistics using RUNSTATS. Number of sorts that were necessary to execute the statement SQL Sorts You can use this value to determine whether new indexes are needed. Use EXPLAIN to check whether and which indexes were used when the selected statement was executed.
September 2011
107
4 Performance
If no hits are found, the result set is empty, and nothing is displayed. End of the note. The following functions are available for further actions: EXPLAIN To display a detailed performance analysis, you can display the access plan for the SQL statement by choosing the EXPLAIN pushbutton. For more information, see The EXPLAIN Function [page 43]. Index Advisor To improve the performance of a query, you can retrieve recommendations about useful indexes using the Index Advisor pushbutton. In addition, you can design new virtual indexes that can be validated before they are actually created. For more information, see The Index Advisor [page 50]. Source To display the ABAP source code from where the SQL statement was executed, you can choose the Show Source pushbutton. Note The Show Source pushbutton is only available if the following applies: o o The SAP system on which the currently monitored database is running is an ABAP system. The SAP system has the minimum Support Package level as described in SAP Note 1493490.
108
September 2011
4 Performance
Common
On the Common tab page, you can find details about the statements and rows. These details include, for example, information about the following: Package cache ID Insert time stamp Partition Authorization ID Buffer pool quality Executions Compilations Rows deleted, inserted, and read
Resource Consumption
The Resource Consumption tab page displays details about the total resource consumption, the proportional resource consumption, and the sorts. These details include, for example, information about the following: Execution time User CPU time System CPU time Worst and best preparation time Rows read and written Average time per row read and per row written Sort overflows Total sort time
I/O
The I/O tab page displays details about the buffer pool, temporary data, and XML storage, that is, about the number of logical and physical reads.
Statement Text
The Statement Text tab page displays the complete text of the SQL statement.
September 2011
109
4 Performance
110
September 2011
4 Performance
Features
The following critical activities are available: Threshold Violations You can use the information on this screen to review violations of defined thresholds. Threshold Configuration You can use the information on this screen to maintain thresholds on the database system.
September 2011
111
4 Performance
Violation Time Time when the threshold violation occurred Member Database member where the violation occurred Predicate that was violated Note Predicate Activities with the value MANUAL were captured manually on the Applications screen. For more information, see Snapshots: Applications (Only DB2 V9.7 or Higher) [page 94] or Snapshots: Applications (Only DB2 V9.5 or Lower) [page 100]. End of the note. Violated Value Value that was exceeded and that violated the threshold predicate Service Superclass Service Subclass Name of the service superclass in which the violation occurred
By default, the history of a threshold violation is kept for two weeks. To delete all recorded threshold violations, choose the Reset Violation History pushbutton. For more information about how to change the size of the violation history, see Data Collection Framework: Collector Configuration [page 227].
112
September 2011
4 Performance
For nested statement calls - for example, during the execution of stored procedures more than one SQL statement can be available. You can retrieve additional information about the bind values and additional details about a single SQL statement as follows: o Retrieve a newly generated access plan by choosing an SQL statement from the list and then the EXPLAIN pushbutton. The new access plan is displayed in a separate window. o Retrieve the access plan as it existed in the SQL cache when the threshold violation occurred by choosing the Section EXPLAIN pushbutton The current access plan is displayed in a separate window. Note DB2 V9.5 only To be able to retrieve the access plan using the Section EXPLAIN pushbutton, you have to set the DB2 registry variable DB2_DUMP_SECTION_ENV to ON. For more information, see SAP Note 1227225. End of the note.
September 2011
113
4 Performance
Drop a threshold by selecting a line from the list and choosing the Drop Threshold pushbutton. Note If the WLM event monitors have not yet been created on the database, you can do so on the Collector Configuration screen under Configuration Data Collection Framework in the DBA Cockpit. For more information, see Configuration: Data Collection Framework [page 227]. End of the note.
114
September 2011
4 Performance
September 2011
115
4 Performance
Defines how to interpret the progress list that is shown in the content detail area after you have selected a line in the overview table The following values are possible are: Concurrent Due to concurrent processing, any element in the progress list can be updated at any time. Processing Mode Serial Interpret the elements in the list as a set of serial phases This means that completed work must equal the total work for element n before the completed work of element n+1 is first updated. This attribute is used to describe the progress of a task that consists of a set of serial phases where a phase must fully complete before the next phase begins. <space> The utility is running as one single phase without parallelism Sequence Number Specifies the current phase the utility is working on Indicates how the utility was started The following values are possible: AUTO Invoker The utility was automatically started by DB2. USER The utility was started manually. Specifies the relative importance of a throttled utility Priority The higher the priority, the more you might experience an impact on overall database performance.
116
September 2011
4 Performance
Shows the current execution state of the utility ERROR The utility is stopped due to an unexpected error situation. State EXECUTE The utility is executing. WAIT The utility is waiting, for example, due to blocked resources or synchronous I/O requests.
September 2011
117
4 Performance
Metrics of the work units The following values are possible: NOT_SUPPORT BYTES Work Metrics EXTENTS INDEXES PAGES ROWS TABLES State Shows the current execution state of the utility phase
118
September 2011
4 Performance
Column
Description Action identifier for a logged operation that depends on the operation run
Operation Type
For more information, see Table 2. OPERATION and OPERATIONTYPE values in section DB_HISTORY administrative view - Retrieve history file information in the IBM DB2 V9.7 Information Center. Identifier for the type of the target object The following values are possible: D
Object Type
Stands for full database P Stands for tablespace T Stands for table
Schema of the target object Name of the target object Additional event information for the operation Number that uniquely identifies an entry in the history file
More Information
For more information about logged events in the database history file, see section DB_HISTORY administrative view - Retrieve history file information in the IBM DB2 V9.7 Information Center at: http://publib.boulder.ibm.com/infocenter/db2luw/v9r7/index.jsp?topic=/com.ibm.db2.luw.sql.rt n.doc/doc/r0022351.html
September 2011
119
4 Performance
120
September 2011
4 Performance
Note If no active inplace table reorganization was found, the system displays the following message: No Inplace Table Reorganizations are running. Only the REORG activities since the database start are displayed. REORGs that were active before the database start are not displayed. End of the note. Depending on your requirements, you can customize the view on the Inplace Table Reorganizations screen using the following functions: Pushbutton Active Only Only inplace table reorganizations with the status Started or Paused are displayed. Choose View All All inplace table reorganizations are displayed including those with status Completed or Suspended. Since DB Start Only inplace table reorganization that have been started after the last restart of the database manager are displayed. Choose Data Source Incl. History File This option additionally reads the DB2 history file. Thus, the data of inplace table reorganizations that were started before the last restart of the database manager is also retrieved. Function
Activities
You can perform the following actions for an inplace reorganization: Pause Select a running inplace reorganization and choose Pause. Resume Select a paused inplace table reorganization and choose Resume. Suspend Select any inplace table reorganization and choose Suspend. As a result of any of these actions, the list of active inplace table reorganizations is refreshed.
September 2011
121
4 Performance
Average physical read time Avg. Phys. Read If you have chosen Total Day, this is the average of all measured average Time (ms) read times. If you have chosen Peak, this is the worst measured read time. Average physical write time Avg. Phys. Write If you have chosen Total Day, this is the average of all measured average Time (ms) write times. If you have chosen Peak, this is the worst measured write time. Data Logical Reads Number of read accesses to data in the buffer pool Number of read accesses to data on disk (I/O) Data Physical Reads The value includes the number of physical reads that were performed synchronously (by the database manager agents) and asynchronously (by the I/O servers). Number of write accesses to data on disk (I/O) Data Physical Writes The value includes the number of physical writes that were performed synchronously (by the database manager agents) and asynchronously (by the I/O cleaners).
122
September 2011
4 Performance
Description Number of read accesses to index data in the buffer pool Number of read accesses to index data on disk (I/O)
Since index data is read only by the database manager agents, this value includes the number of synchronously read index pages. Number of write accesses to index data on disk (I/O)
The value includes the number of physical writes that were performed synchronously (by the database manager agents) and asynchronously (by the I/O cleaners). Total number of COMMIT statements that have been attempted Total number of ROLLBACK statements that have been attempted
COMMIT Statements
ROLLBACK Statements
Automatic ROLLBACKs caused by error situations or deadlocks are not included. Total number of times that applications or connections waited for locks Total elapsed time in milliseconds waited for a lock Total number of deadlocks that have occurred Number of times that locks have been escalated from several row locks to a table lock Number of times that locks have been escalated from several row locks to one exclusive table lock, or the number of times an exclusive lock on a row caused the table lock to become an exclusive lock
If you choose Total Day in the Workload field in the Selection area, the total workload of the relevant day is displayed. The value displayed is calculated using formula maximum value minimum value. Database restarts are taken into consideration. If you choose Peak in the Workload field in the Selection area, the maximum of all measured values is displayed. You can display details for a specific day by clicking a field or selecting a row. A detail area appears with the following tab pages: Snapshot The measured values of the selected day are displayed.
September 2011
123
4 Performance
Interval The delta values of the measurements, which are provided under Snapshot, are displayed. If you select a particular day, snapshots of the database activity are displayed in a two-hour cycle. Note If the database is restarted during one day, the interval displayed after the restart does not equal the delta of two measurements because the counter was reset during the restart. In this case, the absolute value of the last measurement is displayed. End of the note.
124
September 2011
4 Performance
Number of read accesses to tables that resulted in overflow pages, that is, to records that were swapped from their original page. Overflow Accesses Note If there is a high number of overflow accesses in comparison to total read accesses, the table is a candidate for reorganization. End of the note. Page REORGs Note If you want to display history data that is cumulated by day, week, or month, choose the relevant option from the dropdown list in the Statistics field in the Selection area. End of the note. Number of internal page reorganizations during INSERT operations
September 2011
125
5 Space
5 Space
The following sections provide information about the main task areas that are available under Space on the Database tab page of the DBA Cockpit: Overview Automatic Storage Database Tablespaces Containers File Systems Tables and Indexes Single Table Analysis
126
September 2011
5 Space
Field Used Space Minimum Free Space in a Tablespace Maximum Used Space in a Tablespace Database Partitions
Description Used space of all tablespaces as a percentage Free space of the tablespace with the lowest amount of free space in KB Used space of the tablespace with the highest fill level as a percentage
Number of database partitions Total Number The value displayed is only higher than 1 if you are using a multi-partition database.
The job to collect the database and tablespace history and the job to collect the history of tables and indexes are triggered by the standard performance collector job SAP_COLLECTOR_FOR_PERFMONITOR. You can display the schedule of these two jobs in the DBA Planning Calendar by choosing Jobs DBA Planning Calendar on the Database tab page of the DBA Cockpit. From the Category dropdown list, you choose DB Collectors. The default setting is All Actions. Caution Calculating table values with outdated statistics can result in inaccurate values. To calculate update statistics including the calculation of table sizes, use the DBA Planning Calendar [page 234]. End of the caution.
September 2011
127
5 Space
128
September 2011
5 Space
Delete a storage path as follows: 1. Select a storage path and choose the Delete pushbutton. 2. Choose the Execute pushbutton. Note Up to and including DB2 V9.5, you can only delete rows that you had previously added. End of the note. In the Generated SQL Statement area, all the generated SQL statements to be executed are displayed. This area is automatically filled and refreshed as soon as you apply changes to the storage path list. To switch back to display mode, choose the Display pushbutton. Note As of DB2 V9.7, you can enable a database, which was originally created without automatic storage, for automatic storage management by adding storage paths on the Automatic Storage screen. This means that after having created the storage path(s), you can convert nonautomatic storage tablespaces to automatic storage tablespaces as described in Space: Tablespaces [page 131]. End of the note.
September 2011
129
5 Space
Features
The Database screen provides the following: A selection area where you specify filter criteria for the monitoring data An overview table displaying the retrieved monitoring data and providing the space consumption at the end of the selected time period An option to break down all history data that was collected for the selected monitoring data during the specified time frame
Process
To analyze the data provided on this screen, you proceed as follows: 1. You identify the time period where you expect major changes in space consumption. 2. In the Selection area, you specify the appropriate time frame. 3. In case of a multi partitioned system, you can do one of the following: o o To retrieve a general snapshot, you set the partition selection to All and you do not drill down by partitions or tablespaces. To retrieve a more detailed snapshot, you drill down the selection by partition or by tablespace. Note If you want to drill drown by tablespace, you can restrict the result set by limiting it to special tablespaces. End of the note. 4. To refresh the monitoring data, you choose the Apply Selection pushbutton. 5. To display detailed history data, you select a line in the overview table. The data is displayed in the History Details content area. 6. To isolate the time frame during which the problem occurred and to analyze the space growth in more detail, you use the data displayed in the History Details content area.
130
September 2011
5 Space
September 2011
131
5 Space
DMS/SMS Tablespaces
Regardless whether you have chosen automatic storage management tablespaces or manual maintenance of DMS/SMS tablespaces during the SAP system installation, the following information is displayed for all DMS/SMS tablespaces that are maintained manually: Column Tablespace Name Partition TS Type Contents TS State KB Total Page Size No. Containers KB Free Name of the tablespace Number of the partition (only displayed in a multi partition database) Type of tablespace, for example, DMS or SMS Contents of tablespace, for example, any data or temporary data Status of tablespace, for example, Normal or Load Pending Total space in KB used by the tablespace Size of a page in bytes Number of containers Total amount of free space Description
High-Water Mark (KB) Indicates the maximum value of used pages reached Percent Used AUTORESIZE Pending Free Pages Used space in relation to the available space Indicates if the tablespace is enabled for automatic resizing Number of free pages that are pending
132
September 2011
5 Space
In addition, the following information is displayed for each selected tablespace: Field Tablespace Maintenance Name Name of the tablespace Name of the partition group where the selected tablespace is defined A partition group defines a set of partitions. Space Total space in KB Total This information is not displayed when creating tablespaces. Fill level of the selected tablespace as a percentage Used This information is not displayed when creating tablespaces. Free space in KB Free This information is not displayed when creating tablespaces. Description
Technical Settings
The following information is displayed on the Technical Settings tab page: Field Description Describes the type of data that is stored in the tablespace, for example, regular data, large objects, temporary user objects, or temporary system objects Displays the page and extent size in KB, and the number of pages to be prefetched Displays if the space of the tablespace containers is managed by, for example, the database (DMS), the file system (SMS), or automatic storage management Space Management by Note If the AUTORESIZE Enabled checkbox is selected, tablespace containers are automatically extended using the file systems where the containers are located. If you are using automatic storage tablespaces, the AUTORESIZE Enabled checkbox is selected by default and cannot be deselected.
Contents
September 2011
133
5 Space
Field
Description If the Reclaimable Storage Enabled checkbox is selected, the tablespace is enabled for reclaimable storage. This means that unused extents can be released to the system for reuse. Tablespaces that have been created with DB2 V9.7 or higher do have this property. End of the note. Displays information about disk performance, such as: Overhead Displays I/O controller overhead and disk seek and latency time in milliseconds. This value is used to determine the cost of I/O during query optimization. Transfer Rate Time to read one page into memory in milliseconds. This value is used to determine the cost of I/O during query optimization. File System Caching
Disk Performance
The value displayed indicates the use of file system caching as follows: o YES File system caching has been explicitly enabled using the FILE SYSTEM CACHING clause of the CREATE or ALTER TABLESPACE statement. o NO File system caching has been explicitly disabled using the FILE SYSTEM CACHING clause of the CREATE or ALTER TABLESPACE statement. o OS No explicit file system caching clause has been specified. Therefore, the default of the file system is used.
Recovery
Dropped tables in the specified tablespace may be recovered using the RECOVER TABLE ON option of the ROLLFORWARD command. By default, the buffer pools are displayed that match the page size of the tablespace. If required, you can add a new buffer pool. For more information, see Maintaining Buffer Pools [page 199].
Buffer Pool
For more information about technical settings, see the IBM documentation SQL Reference.
134
September 2011
5 Space
Storage Parameters
For tablespaces that are completely managed by automatic storage management or that have at least AUTORESIZE enabled, the following values are displayed on the Storage Parameters tab page: Field Settings Initial Size Size Current Size Last Resize Displays the current size Date and time of last automatic resize operation Date and time when the automatic resize operation failed Note Last Resize Terminated with SQL Error This field only appears if the last automatic resize operation failed. End of the note. The SQL error is displayed in the lower half of the Tablespaces screen. Initial space allocated when a tablespace is created Description
Containers
This tab page provides information about the containers of the selected tablespace.
September 2011
135
5 Space
Maintaining Tablespaces
You can maintain tablespaces, for example, Change, Add, or Delete tablespaces. For more information, see Maintaining Tablespaces.
Changing a Tablespace
1. On the Tablespace screen, select a tablespace and choose the Change pushbutton. The Change Tablespace dialog box appears where you can change the following settings: Field Technical Settings Prefetch Size Number of pages to be prefetched I/O controller overhead and disk seek and latency time in milliseconds Overhead You can use this value to determine the cost of I/O during query optimization. Time to read one page in milliseconds Transfer Rate You can use this value to determine the cost of I/O during query optimization. Name of the allocated buffer pool Note Buffer Pool You can enter only buffer pools that match the page size of the tablespace. End of the note. Description
136
September 2011
5 Space
You can activate or deactivate file system caching. Note File System Caching As long as neither the FILE SYSTEM CACHING nor the NO FILE SYSTEM CACHING clause has been specified, the default caching options of the underlying file system type are used. The first use of any of these options explicitly sets this value. You cannot fall back to the default behavior. End of the note. Dropped Tables Storage Parameters AUTORESIZE Enabled Enable this option if you want to allow DB2 to automatically enlarge tablespace containers using the file system where the containers are located. You can enter one of the following values: o NONE There is no maximum size limit. In this case, you allow DB2 to extend containers until they occupy all file systems where the containers are located o Maximum Size An absolute value in KB This value defines an upper threshold that shall not be exceeded by automatic extensions Note This field is valid only if the AUTORESIZE Enabled option is selected. End of the note. Size in KB or in percent by which the tablespace is extended if it has become full You can specify an absolute value in KB or a relative percentage. Increase Size Note This field is valid only if the AUTORESIZE Enabled option is selected. End of the note. Selecting this checkbox enables to you recover dropped tables using the RECOVER TABLE ON option of the ROLLFORWARD command.
September 2011
137
5 Space
If a tablespace is not managed by automatic storage management, you can add or delete containers as follows: o To add containers, choose the Add Container pushbutton. The system automatically suggests a default path where the container is located. However, you can modify that path by manually editing the line. Caution Adding or changing containers might result in rebalancing, which has a heavy impact on system performance. Containers (Optionally available) End of the caution. At least one container must be available for each partition. If you are using a multi partition database, you need to add containers for all partitions of the corresponding partition group. If you have to change container sizes, we recommend that you enter an appropriate value in the Resize all containers to field to ensure a balanced distribution of data on the different containers. Caution Different container sizes might result in bad performance of the database. End of the caution. o To delete containers, select one or more lines in the table and choose Delete.
Adding a Tablespace
1. On the Tablespace screen, choose the Add pushbutton. 2. In the Add Tablespace dialog box, specify a name and a partition group. Recommendation We recommend that you use uppercase letters for the tablespace name. Using lowercase letters or special characters makes accessing the selected tablespace with the DB2 command line processor less comfortable. End of the recommendation. 3. Enter the technical settings. By default, the system displays SAP's recommendations. 4. Add containers. If you are using a multi partition database, you must add containers for all partitions of the corresponding database partition group.
138
September 2011
5 Space
Caution This step does not apply to tablespaces managed by automatic storage management. End of the caution. 5. To confirm your entries, choose the Execute pushbutton.
Deleting a Tablespace
On the Tablespace screen, select a tablespace and choose the Delete pushbutton. Caution You cannot delete tablespaces that are still used by the SAP system, that is, if they are related to some data class. You must delete the data class before deleting the tablespace. End of the caution.
September 2011
139
5 Space
Rebalancing a Tablespace
Regular and large tablespaces that are managed by automatic storage management have to be rebalanced if the following applies: You created new containers on recently added storage paths. A container and its content has to be distributed to the remaining containers. If a database was converted to automatic storage by adding storage paths, the tablespaces still reside in the old containers and do not yet have containers in the automatic storage database. For the first two cases, the rebalancing process is automatically started. If a database was converted to automatic storage, however, you have to manually start the rebalancing process by selecting a tablespace in the overview table and choosing the Rebalance pushbutton. After the rebalancing process has been started, data is moved from the containers of the dropped storage paths to the containers on new storage paths. The rebalancing process runs asynchronously in the background and does not affect the availability of data. To monitor the rebalancing process, check the TS State column in the overview table. If this column shows any of the values Reverse Rebalance, Forward Rebalance or DMS rebalancer is active, select the TS State column to see a detail view of the rebalancing progress. For each partition, the following data is displayed: Column Partition Tablespace Name Partition number Name of the tablespace The rebalancer mode can be one the following No Rebalance The rebalancer process has finished on this partition. Reverse Rebalance Mode Data movement starts at the high-water mark extent and moves in reverse order through the tablespace, ending with the first extent in the tablespace. Forward Rebalance Data movement starts with the first extent in the tablespace and ends with the high-water mark extent. Description
140
September 2011
5 Space
Extents Processed Extents Remaining Last Extent Moved Priority Restart Time Start Time Progress (%)
Number of tablespace extents that have already been moved to the new location Number of tablespace extents that still need to be moved
Number of the last extent that was moved Priority with which the rebalancing process is running in the database Time when the rebalancing process was restarted after it had been paused or stopped Time when the rebalancing process was first started Displays the progress of the reduction process in percent
More Information
Configuration: Data Classes [page 224]
September 2011
141
5 Space
To display detail information about a container, choose a line in the overview table. For more information, see Space: Tablespaces [page 131].
Maintaining Containers
You can maintain tablespace containers by selecting a line in the table on the Container screen and choosing the Change, Add, or Delete pushbutton. For more information about how to maintain container settings, see Maintaining Tablespaces [page 136].
File System Name KB Total KB Used Percentage Used KB Free Percentage Free
142
September 2011
5 Space
As of DB2 V9.1 Fix Pack 6, the following information is displayed: Column Partition Mount Point KB Total KB Used Percentage Used KB Free Description Number of the partition (only displayed if you are using a multi partition database) Location in the OS directory structure where the file system appears Total size of the file system in KB Total amount used of the file system in KB Used percentage of total size of the file system Total amount free of the file system in KB
Percentage Free Free percentage of total size of the file system in KB File System Type Device Name Block Size File system type Name of the device Block size of the file system
September 2011
143
5 Space
Features
The following subset of task areas is available: Top Space Consumers You can use the information on this screen to identify the top space consumers in your database for a specified time period. REORG Candidates You can use the information on this screen to identify tables that have been recommended for a reorganization. Compression Status You can use the information on this screen to check the compression status of a table or if a table is a suitable candidate for compression. Virtual tables You can use the information on this screen to check whether or not a table can be virtualized or materialized to save space. Indexes You can use the information on this screen to obtain an overview of all indexes in your database.
144
September 2011
5 Space
Features
The Top Space Consumers screen provides the following: A selection area where you specify filter criteria for the collected data An overview table displaying the retrieved history data This overview table also includes a subset of space-relevant data reflecting the last set of values that was collected during the selected time period. An option to break down all history data that was collected for the selected monitoring data during the specified time frame.
Process
To be able to analyze the space situation in your database, you proceed as follows: 1. You identify the time period where you expected a major impact on space consumption of your tables, for example: o o o Time periods with archiving and REORG activities Mass-processing of application data Installation of new application functions
2. In the Selection area, you specify the appropriate time frame. Note If you are using a multi partition database system and you want to retrieve a general snapshot, you set the partition selection to All and you do not filter by partition. To retrieve a more detailed snapshot, you can drill down the selection by partitions. If a single partition or a subset of partitions shows deviations compared to other members, you can also restrict the partition selection. Deviations are usually an indication for a bad partitioning key. End of the note. 3. To refresh the result set, you choose the Apply Selection pushbutton.
September 2011
145
5 Space
4. To display detail information about the growth of a table over time, you select the appropriate entry in the overview table. The data is displayed in the History Details area. 5. To display further details of the selected table, you select a line in the overview table and choose the Detailed Table Analysis pushbutton. You are directed to the Space Single Table Analysis screen. For more information about the data provided on the Single Table Analysis screen and how to evaluate it, see Analyzing and Maintaining a Single Table [page 154].
REORG Table Indicates if a table reorganization is required. REORG Index Indicates if only an index reorganization is required. Check Date Check Time Date of last REORGCHK Time of last REORGCHK
146
September 2011
5 Space
Compressed Tables
On this tab page, all tables meeting at least one of the following requirements are displayed: The table is enabled for data compression and contains compressed records. The indexes of a table are enabled for index compression and are compressed. Column Table Schema Table Name Data Compression Index Compression Name of the table schema Name of the table If the table is enabled for data compression, this checkbox is selected. If the index of the selected table is enabled for index compression, this checkbox is selected. Total savings of space Savings (MB) This includes all savings regardless, whether they are caused by data or index compression Total size of the table Total Size The value displayed is the sum of DATA, LONG, LOB, XML, and index pages over all database partitions, Percentage of savings Description
Savings (%)
September 2011
147
5 Space
Description
Compression Candidates
On this tab page, tables that meet at least one of the following requirements are displayed: A table has already been compressed but more space could be saved by recompressing it based on a better compression dictionary. A table has been enabled for data compression but still contains compressed records that are not yet compressed. A table whose indexes are not yet enabled for index compression A table whose indexes are compressed but more space could be saved by recompressing the table Column Table Schema Table Name Name of the table schema Name of the table Description
Data Compression If row compression is enabled, this option is selected. Index Compression If index compression is enabled, this option is selected. Total savings of space that can be achieved This includes all savings regardless, whether they are caused by data or Estimated Savings index compression. (MB) If the table has already been compressed, then the value displayed does not represent the new total savings but the additional savings. Total Size Total size of table if the compression would be performed
Estimated Savings Percentage of estimated (additional) savings (%) Last Check Date Last Check Time Last check date Last check time
148
September 2011
5 Space
Total Compression Ratio Total compression ratio in percent for this table Data Total amount of savings in KB for the data of this table
Data Compression Ratio Total compression ratio in percent for the data of this table Index Total amount of savings for all indexes of this table
September 2011
149
5 Space
Index Compression Ratio Estimated Savings Total Total Compression Ratio Data Data Compression Ratio Index Index Compression Ratio Additional Savings after Compression Check Date Check Time
Total amount of estimated savings in KB of this table (including indexes) Estimated total compression ratio in percent Amount of estimated savings of data in KB Estimated data compression ratio in percent Amount of estimated savings in KB for all indexes Estimated index compression ration Total amount of savings in KB if the table would be compressed again Date when table was last checked Time when table was last checked
3. To schedule the compression job, choose OK. The appropriate REORG jobs are scheduled in the DBA Planning Calendar where you can view the progress of the compression job. Note Depending on the required actions to perform the compression, there can be separate REORG jobs for table and for indexes. End of the note.
150
September 2011
5 Space
3. To schedule the compression job, choose OK. A job is scheduled in the DB2 administrative task scheduler (ATS). You can view view the progress of the compression check job either under Jobs Scheduled DB Tasks or in the DBA Planning Calendar where you choose DB Collectors from the Category dropdown list.
September 2011
151
5 Space
Candidates for Virtualization Displays a list of tables that are candidates for being dropped and re-created as virtual tables. If you choose the Convert Empty Tables pushbutton, a background job is scheduled that checks each table for the following: o o o o Whether it is empty Whether it is not volatile Whether it does not have a partitioning key Whether it does not use MDC tables
Tables that meet these conditions are dropped and re-created as virtual tables. Note The use of virtual tables is transparent to the ABAP Dictionary. End of the note. Caution Before you drop tables and re-create them as virtual tables, make sure that you have read SAP Note 1151343. End of the caution.
152
September 2011
5 Space
In the overview table, the following information is displayed: Column Index Schema Index Name Table Schema Table Name Number of Leaves Cardinality Est. Phys. Index Size (KB) Est. Log. Index Size (KB) Last Used Description Name of the index schema Name of the index Name of the table schema Name of the table Number of leaves in the index Cardinality of the index Estimated physical index size in KB Estimated logical index size in KB Date when the index was last used
September 2011
153
5 Space
Features
The Single Table Analysis screen provides the following areas: A selection area, where you specify the schema and name of the table to be analyzed A summary area that provides an extract of important key figures and status information of the specified table Tab pages with detail information that is relevant for tables or indexes In the summary area, the following information is displayed: Field Size Total table size Total Table Size This is the summary of all DATA, LONG, LOB, and XML pages over all database partitions Total size of all indexes over all database partitions Number of indexes defined on the selected table Description
154
September 2011
5 Space
Status of the statistics The following values are possible: Statistics available RUNSTATS Status No Statistics available VOLATILE Table Without Statistics Unwanted Statistics for VOLATILE Table REORG Status Indicates whether and how a REORG should be performed Indicates whether and how the table is compressed Note A table or index is assumed to be compressed if the following applies: Compression Status In the system catalog, the table or index is enabled for compression. The data or index pages have been compressed by performing an appropriate reorganization. End of the note. Savings by Compression Note The following fields are only relevant if the compression status is accordingly. That is, if the Compression Status tab page displays that a table or at least its index is compressed. End of the note. Current percentage of all savings for row and index compression compared to the total size of all DATA, LONG,LOB, XML, and index objects
September 2011
155
5 Space
Current total savings of space Current Savings This value includes all savings regardless whether they were caused by data or index compression. Compression ratio that would be reached after compression Estimated Compression Ratio Note This value is only displayed if a compression check has already been performed. End of the note. Estimated Savings Additional Savings after Compression Savings that would be reached after a compression Additional savings after compression for tables that have already been compressed
The following tab pages are available on the Tables and Indexes Details screen providing detail information about tables or indexes: System Catalog Table Structure Indexes This tab page contains an overview list of all indexes and provides integrated tab pages with detail information about, for example, system catalog, structure and status of the index, as well as information about the compression status and REORGCHK. Table Status Compression Status REORGCHK
Activities
To optimize tables and indexes, you can use the following pushbuttons on the Single Table Analysis screen: RUNSTATS Schedules a RUNSTATS job for a single table using the DBA Planning Calendar. For more information, see Updating Statistics [page 249].
156
September 2011
5 Space
Note To decide if statistics are out-of-date, you can check the statistics time. Alternatively, you can count the number of entries in the table by choosing the Count pushbutton to compare the real number of records with the Cardinality field on the System Catalog tab page. End of the note. REORG Schedules a REORG job for the table using the DBA Planning Calendar. For more information, see Reorganizing Tables [page 252]. Compression Check Schedules a job that checks whether a table would benefit from compression Caution The Compression Check job is expensive and can considerably decrease the system performance. We, therefore, recommend that you schedule this job at a time with a low system workload and not in dialog mode. End of the caution. Compression On/Off Switches the compression status of a table or index. Note The term compression is used here as a synonym for both data and index compression. End of the note. The Compression Options dialog box appears displaying the current compression status of the table, that is, if compression is enabled or disabled, and if a compression dictionary exists. Depending on the status, you can choose between the appropriate actions, such as: Action Enable Compression Description Switches the compression attributes for the table and its indexes on
Enable Compression and Switches the compression attributes on and schedules a REORG Run REORG job that compresses the table data and its indexes Run REORG to Remove Schedules a REORG job with the RESET DICTIONARY option for a Dictionary table for which the compression attribute was switched off Disable Compression Switches the compression attributes off
September 2011
157
5 Space
Disable Compression Switches the compression attributes off and schedules an REORG that and Run REORG uncompresses table and index data Schedules a REORG job that rebuilds a dictionary on which compression is based for an already compressed table Run REORG to Rebuild Dictionary This is recommended if a new compression check indicates a better compression ratio or if the compression attributes were switched on but the table had not yet been reorganized.
Schedules a REORG job for a table where the compression attributes Run REORG to Build were switched on but the table and indexes have not yet been Dictionary compressed. Note Before you enable compression, we recommend that you check the possible savings by starting a compression check using the Compression Check pushbutton. End of the note. VOLATILE On/Off Changes the VOLATILE attribute of a table Note If you set the VOLATILE attribute, the table is no longer processed by automatic RUNSTATS. End of the note.
158
September 2011
5 Space
September 2011
159
5 Space
Statistics Data Last RUNSTATS Cardinality Date and time of the RUNSTATS Number of data records in the table Number of rows that have been counted by a SELECT(*) statement Counted Rows This information is only displayed if you choose the Count pushbutton. Deviation of the number of rows provided by RUNSTATS in the system catalog from the number of rows provided by a SELECT COUNT(*) statement This information is only displayed if you choose the Count pushbutton. Number of records that have overflowed Overflow Records Records overflow when a data record is updated and the new data record is larger than the old one, or when a column is added to a table. The percentage of overflow records is used as a decision criteria for REORG (see F1 formula on the REORGCHK tab page) Number of pages containing data Total number of pages in the table If a RUNSTATS was executed using the PROFILE option, this profile is displayed. The profile is the same as the RUNSTATS command that was executed with the SET PROFILE option. Note RUNSTATS Profile If a profiles exists, it is used by automatic RUNSTATS and therefore the resulting statistics can deviate from the default setting for statistics. In addition, if you manually execute a RUNSTATS, you have to make sure that you update statistics by using the profile or by explicitly choosing another type of statistics. End of the note. Compression Value Compression Row Compression Indicates whether value compression is enabled for the table Indicates whether the table is enabled for row compression
Deviation
160
September 2011
5 Space
Average Length of Compressed and Average length of all rows (compressed and uncompressed) in Uncompressed Rows From bytes Total Number of Rows Percentage of Compressed Compressed rows as a percentage of the total number of rows in Rows From the Total the table Number of Rows Percentage of pages saved by compression Approximate Percentage of Pages Saved Note Only data pages are taken into account. End of the note.
The following information is only displayed for the local system and if the table is defined as a transparent table in the ABAP Dictionary: Column SAP Column Name SAP Key SAP Type Description
Column is part of the primary key defined in the ABAP Dictionary. Data type of the column defined in the ABAP Dictionary
September 2011
161
5 Space
Column
Description
If the statistics for the table have been gathered using the WITH DISTRIBUTION option, the following information is displayed: Column Name Name of the column to which the statistics apply If the Type column contains the value F, the value n in this column identifies the nth most frequent value. If the Type column contains the value Q, n in this column identifies the nth quantile value. Data value as a character literal If the Type column contains the value F, this is the number of occurrences of the Value column. If the Type column contains the value Q, this is the number of rows whose value is less than or equal to the value in the Value column. If the Type column contains the value Q, this is the number of distinct values that are less than or equal to the value in the Value column. Specifies how to interpret the displayed values The following values are possible: F Type Indicates the frequency value Q Quantile value
Seq. No
162
September 2011
5 Space
September 2011
163
5 Space
Field XVIL
Description
Index over XML column (logical) XVIP Index over XML column (physical) Statistics Time Date and time of last RUNSTATS
Together with the summary information, there are detail tab pages displaying all information that is related to the currently selected index. By default, the data of the first index is displayed. To view the details of another index, select a table from the list. Note If there is only one index, the summary list does not appear. End of the note.
164
September 2011
5 Space
Technical Attributes Percentage of free space reserved in the index pages Free Space Reserved This free space is taken into account by the LOAD and REORG utilities of DB2. Note Compression This field is only available as of DB2 V9.7. End of the note. Indicates whether the index is enabled for compression Reverse Scans Supported Statistics Data Last RUNSTATS Date and time of last RUNSTATS Indicates whether the index supports reverse scans
Number of Leaves Number of index leaves Number of Levels Number of index levels Sequential Pages Number of index leaves that are physically located on the hard disk and that are sorted by index without large intervals between them Relative density of the sequential pages as a proportion of the total number of index pages 100% is the optimum value. Degree of index fragmentation Cluster Ratio A value of 100% means no fragmentation and this is the optimal value. Finer measurement of the degree of clustering Cluster Factor If statistics are not collected or if the index is defined on a nickname, a value of -1 is displayed. Cardinality of the index Cardinality This value might be different from the cardinality of the table for indexes that do not have a one-to-one relationship between the table rows and the index entries. Number of different values in the first column of the index
Density
September 2011
165
5 Space
First 2 Key Cardinality First 3 Key Cardinality First 4 Key Cardinality Full Key Cardinality
166
September 2011
5 Space
Note The following information is only displayed for the local system and if the table is defined as a transparent table in the ABAP Dictionary: Column SAP Column Name SAP Type SAP Length End of the note. Description Name of the column defined in the ABAP Dictionary Data type of the column defined in the ABAP Dictionary Length of the column defined in the ABAP Dictionary
September 2011
167
5 Space
Partitioning characteristic of the index The following values are possible: N Index Partitioning Non partitioned index P Partitioned index Blank Index does not reside on a partitioned table. Index Requires Rebuild
Indicates whether the index is using large row IDs (RIDs), that is, 4-byte page number, 2-byte slot number The following values are possible: Yes The index is using large RIDs. Large RIDs No The index is not using large RIDs Pending The table where the index is defined supports large RIDs, that is, the table resides in a large tablespace. The index for the table or data partition, however, has not yet been reorganized or rebuilt. Therefore, the table is still using 4-byte RIDs, and the table or index has to be converted to use large RIDs. Physical index format The following values are possible: Yes Index Compressed Index is in compressed format No Index is in uncompressed format Note
168
September 2011
5 Space
If the physical index format does not match the compression attribute, you have to reorganize the index to convert the index to the defined format. End of the note. Logical Size Physical Size Index ID Note If the index is defined on a table that is partitioned or using data partitioning, the index status is separately available for each partition or data partition. End of the note. Amount of disk space in KB that is logically allocated for the index Amount of disk space in KB that is physically allocated for the index Identifier for the index
September 2011
169
5 Space
Physical index format Yes Index is in compressed format. Index Compressed No Index is in uncompressed format. If the physical index format does not match the compression attribute, an index reorganization is required to convert the index. Note This value is the same for each entry of an index or index partition for each database partition in a DPF environment. End of the note. Saved Pages If the index is not physically compressed (that is, INDEX_COMPRESSED is N), the displayed value represents the estimated percentage of leaf pages saved, as if the index were actually compressed. If the index is physically compressed (that is, INDEX_COMPRESSED is Y), the displayed value reports the PCTPAGESSAVED value from the system catalog view (either SYSCAT.INDEXES or SYSCAT.INDEXPARTITIONS). For more information, see Indexes: System Catalog [page 164].
Note This value is the same for each entry of an index or index partition for each database partition in a DPF environment. End of the note. If the index is not physically compressed (that is, INDEX_COMPRESSED is N), the displayed value represents the estimated number of leaf pages saved as if the index were actually compressed. If the index is physically compressed (that is, INDEX_COMPRESSED is Y), the displayed value reports the calculated number of leaf pages saved. This calculation is based on the PCTPAGESSAVED and NLEAF values from the system catalog view (either SYSCAT.INDEXES or SYSCAT.INDEXPARTITIONS). If either PCTPAGESSAVED or NLEAF are invalid values (-1), this value is also set to -1. For more information, see Indexes: System Catalog [page 164]. Time of Collection Date and time of the last compression check for this index
170
September 2011
5 Space
Note If the index is defined on a table that is partitioned or using data partitioning the index status is available for each partition or data partition separately. End of the note.
Number of deleted tree leafs in relation to total tree leafs of index tree
Number of index entries in the index Cardinality For some indexes, this value can be different from table cardinality. For example, the index cardinality on XML columns might be higher than the table cardinality.
Number of PseudoNumber of pseudo-deleted RIDs. Deleted RIDs Note As of DB2 V9.7, the REORGCHK information for indexes of tables with data partitioning is available per data partition. In this case, the data is displayed in list format.
September 2011
171
5 Space
Long Objects
LOB Objects
Amount of disk space in KB that is logically allocated for the table Data Objects For MDC tables, this size includes the logical size of the block map object. This size represents the logical size of the base table only. Space that is consumed by LOB data, long data, indexes, and XML objects is reported by other fields described in the following. Amount of disk space in KB that is logically allocated for long field data in a table Amount of disk space in KB that is logically allocated for long field data in a table
Long Objects
LOB Objects
172
September 2011
5 Space
Description Amount of disk space in KB that is logically allocated for XML data in a table Amount of disk space in KB that is logically allocated for the indexes
Current status of an inplace table reorganization on the table The following values are possible: ABORTED The inplace table reorganization has status PAUSED but is unable to resume. A STOP is required. For more information, see Utilities: Inplace Table Reorganization [page 120]. EXECUTING Inplace REORG Status The inplace table reorganization is currently being executed. NULL This value only appears if no inplace reorganization has been performed on the table. PAUSED The inplace table reorganization currently pauses and can be resumed again. For more information, see Utilities: Inplace Table Reorganization [page 120]. Indicates whether a REORG is pending for the table REORG Pending This state is possible for ALTER operations like DROP COLUMNthat require an offline reorganization of the table Number of ALTER operations after which a reorganization is required and that have been performed against this table since the last reorganization For an MDC table in a DMS tablespace, this value indicates the amount of disk space that can be reclaimed by running the REORG command with the RECLAIM option. For any other table, the value is zero. Indexes Require Rebuild
September 2011
173
5 Space
Other Technical Attributes Indicates if a table is using large row IDs (RIDs) (4-byte page number, 2-byte slot number) The following values are possible: YES The table is using large RIDs. Large RIDs NO The table is not using large RIDs. PENDING The table supports large RIDs (that is, the table is in a large tablespace) but at least one of the indexes for the table has not yet been reorganized or rebuilt. Therefore, the table is still using 4 byte RIDs, which means that the table or indexes have to be converted. Indicates whether the table is using large slots (which allows more than 255 rows per page) The following values are possible: YES The table is using large slots. Large Slots NO The table is not using large slots. PENDING The table supports large slots (that is, the table resides in a large tablespace) but an offline table reorganization or a table truncation operation has not yet been performed. Therefore, the table is still using a maximum of 255 rows per page. Blocks Pending Cleanup Indicates the number of blocks pending cleanup for MDC tables. For tables that do not use MDC, this value is always 0.
174
September 2011
5 Space
Indicates how the statistics were generated The following values are possible: System fabricated Statistics are gathered by the system without a table or an index scan. These statistics are stored in-memory and are different from the statistics that are stored in the system catalog. This is a temporary state and finally full statistics are gathered by DB2 and are stored in the system catalog. System asynchronously gathered Type of Statistics Statistics are gathered asynchronously by the system. The statistics have been collected automatically by DB2 by a background process and stored in the system catalogs. System synchronously gathered Statistics are gathered synchronously by the system. User gathered Statistics are gathered by the user. Undef Unknown type of statistics or information that is not available for the current database release Current Dictionary Size
Indicates the type of indexes that are currently in use for the table The following values are possible: 1 Index Type Appears if type-1 indexes are being used 2 Appears if type-2 indexes are being used
September 2011
175
5 Space
Availability Describes the status of the table The following values are possible: No The table is not available and all other output information that relates to the size and state is 0. Available YES The table is available. Note Rollforward through an unrecoverable load makes a table unavailable. End of the note. Read Access If the table is read-only, the value is YES. Otherwise, the value is NO. Only No Load Restart The value YES indicates that the table is in partially loaded state that does not allow a load restart. Otherwise, the value NO is returned.
If the table is partitioned or using data partitioning the table status is available for each partition or data partition separately. In this case, an overview list that contains a subset of the information described before, is displayed first. To view the full details, you can select an entry in the list. By default the first entry is displayed. Note The physical sizes returned consider full extents allocated for the appropriate object and include the Extent Map Page (EMP) extents for objects created in DMS tablespaces. The logical size is the amount of space that is known for this table. It might be less than the amount of space that is physically allocated to hold object data for the table, for example, in case of a logical table truncation. The logical size returned considers full extents that are logically allocated for the object and, for objects created in DMS tablespaces, an estimate of the EMP extents. End of the note.
176
September 2011
5 Space
September 2011
177
5 Space
Field REORG
Description
The dictionary was automatically built by the REORG RESETDICTIONARY utility. TABLE GROWTH The dictionary was built in the course of INSERT operations. Build Time Rows Sampled Estimated Compression Saved Pages Estimated amount of pages in percent that will be saved after compression Estimated amount of bytes in percent that will be saved after compression Estimated size of compression dictionary in bytes if the table will be compressed Size of the expansion dictionary measured in bytes Expansion Dictionary Size If a historical dictionary exists, this value is the sum of the current and historical dictionary sizes. Average length of rows if the table will be compressed Date and time when the dictionary was built Rows that have been sampled to calculate the compression results
Saved Bytes
Dictionary Size
In the following cases, compression information is available more than once: The table is distributed across several database partitions. The table uses data partitioning The table contains XML data that is compressed separately from normal DATA pages. If any of these conditions is fulfilled, an overview list containing a subset of the information described before, is displayed first. To view the full details, you can select an entry in the list. By default, the first entry is displayed.
178
September 2011
5 Space
Number of records that have overflowed Number of Overflow Records overflow when a data record is updated and the new data Pages record is larger than the old one, or when a column is added to a table. Note As of DB2 V9.7, the REORGCHK information for tables with data partitioning is available per data partition. In this case, the data is displayed as list. End of the note.
September 2011
179
Log Files
180
September 2011
7 Configuration
7 Configuration
The following sections provide information about the main task areas that are available under Configuration on the Database tab page of the DBA Cockpit: Overview Database Manager Database Registry Variables Parameter Check Configuration History Buffer Pools Workload Management Database Partition Groups Special Tables Regarding RUNSTATS Automatic Maintenance Settings Data Classes Data Collection Framework Monitoring Settings
September 2011
181
7 Configuration
Build Level
PTF
Fix Pack
For the operating system, the following information is displayed: Operating System Host Name Partition OS Name OS Version OS Release Total CPUs Name of the system Partition number Name of the operating system Version number of the operating system Release number of the operating system Total number of physical CPUs of the system Description
182
September 2011
7 Configuration
Description Number of configured physical CPUs of the system Total amount of memory in the system in MB
If the database is distributed over several partitions, the operating system data is displayed in a table. End of the note. If the system has been installed as a high-availability disaster recovery (HADR) system, the following additional information is displayed: HADR Information Description Current HADR connection status of the database The following values are possible: Connect Status CONGESTED CONNECTED DISCONNECTED Local HADR host name Local Host The value is displayed as a host name string or an IP address string, for example, 1.2.3.4. Local HADR TCP service Local Service The value is displayed as a service name string or a port number string. Average of the gap between the primary log sequence number (LSN) and the standby log LSN The gap is measured in bytes. Primary Log File Name of the current log file on the primary HADR database Current log position of the primary HADR database Primary Log LSN The log sequence number (LSN) is a byte offset in the log stream of the database.
Log Gap
September 2011
183
7 Configuration
Page number in the current log file indicating the current log position on the Primary Log primary HADR database Page The page number is relative to the log file, for example, page zero is the beginning of the file. Current HADR synchronization mode of the database The following values are possible: HADR Syncmode ASYNC NEARSYNC SYNC HADR Timeout Number of seconds without any communication from its partner server after which an HADR database server considers that the connection between them has failed Number of missed heartbeats on the HADR connection Heartbeat If the database is in HADR primary or standby role, this field indicates the health of the HADR connection. If the database is in HADR primary or standby role, the meaning of this field depends on the value of the Connect Status field. The following values are possible: CONNECTED Displays the connection time CONGESTED Connect Time Displays the time when the congestion began DISCONNECTED Displays the disconnection time If there has been no connection since the HADR engine dispatchable unit (EDU) was started, the connection status is reported as Disconnected, and the HADR EDU startup time is used for the disconnection time. Since HADR connect and disconnect events occur relatively seldom, the time is collected and reported even if the DFT_MON_TIMESTAMP switch is off. This element should be ignored if the database's HADR role is STANDARD. Host name of the HADR remote host Remote Host The value is displayed as a host name string or an IP address string, for example, 1.2.3.4.
184
September 2011
7 Configuration
Name of the HADR remote instance Remote HADR TCP service This value is displayed as a service name string or a port number string. Current HADR role of the database The following values are possible:
HADR Role
PRIMARY STANDARD STANDBY Current HADR state of the database The following values are possible: DISCONNECTED
HADR State
Name of the current log file on the standby HADR database Current log position of the standby HADR database Log sequence number (LSN) is a byte offset in the log stream of the database. Page number in the current log file indicating the current log position on the standby HADR database The page number is relative to the log file, for example, page zero is the beginning of the file.
September 2011
185
7 Configuration
Information about user groups of the database manager Information about authentications of the database manager and about clients Information about the memory of the database manager Information about agents Information about the database application remote interface (DARI) Information about the configuration of the synchronization manager and the transaction manager Information about the transaction manager Information about network characteristics such as communication protocols
Security Authentication Memory Agents Application Remote Interface Sync Point Manager Transaction Manager Network
Fast Communication Information about the Fast Communication Manager (FCM), that is, the Manager configured communication in a multi-partition database DB2 Discovery Information about the configuration of the discovery mode Single parameters that are not accessible to the groups described above as well as parameters that are not known by the DBA Cockpit, for example, those of a new database release
Others
186
September 2011
7 Configuration
For each node, the following information about the respective parameters is displayed: Column Description
Configuration Parameter Description of the parameter Technical Name Current Value Deferred Value DB2 technical name of the parameter Currently active value of the parameter Value that becomes active after a restart of the database instance
For more information about these parameters, see the IBM DB2 online documentation. For information about maintaining the database configuration parameters, see Maintaining the Database Configuration [page 189].
Information about the memory that is available for the application Information about log files and logging parameters Information about log file management parameters
September 2011
187
7 Configuration
Tree Node Backup & Recovery TSM Locks Space Applications DB2 Data Links Manager High Availability
Description Information about recovery availability and backups Information about Tivoli Storage Management (TSM) Information about locks, for example, the percentage of lock lists per application Information about containers and tablespaces Information about applications that connect to the database Information about the DB2 Data Links Manager (DB2 UDB Version 8 only) Information about the system configuration is only displayed if you are running a high availability system. Single parameters that are not accessible to the groups described above as well as parameters that are not known by the DBA Cockpit, for example, those of a new database release
Others
For each node, the following information about the respective parameters is displayed: Column Description
Configuration Parameter Description of the parameter Technical Name Current Value Deferred Value Note In a multi partition environment, the parameters can vary for each partition. For more information about how to compare the configuration of several database partitions, see Comparing Database Configuration Parameters For Several Database Partitions [page 189]. Depending on your database release level, the tree nodes that are displayed can vary, that is, some might not be visible or others might be added. End of the note. For information about how to maintain the database configuration parameters, see Maintaining the Database Configuration. For more information about these parameters, see the IBM DB2 online documentation. DB2 technical name of the parameter Currently active value of the parameter Value that becomes active after a restart of the database instance
188
September 2011
7 Configuration
Procedure
1. Select the parameter that you want to change and choose the Change pushbutton. The Change Database Configuration Parameter dialog box appears. 2. Enter the new configuration parameter values. Note Some configuration parameters are enabled for automatic value adjustment. In this case, the AUTOMATIC checkbox is displayed. If you select AUTOMATIC, the value is automatically maintained by DB2. End of the note. 3. To check your entries, choose the Check pushbutton. The generated CLP commands that are based on your input are displayed. This area is automatically filled and refreshed whenever you choose the Check pushbutton. 4. To confirm your entries, choose the Execute pushbutton.
September 2011
189
7 Configuration
190
September 2011
7 Configuration
You can access the Parameter Check screen by calling the DBA Cockpit and choosing Configuration Parameter Check on the Database tab page of the DBA Cockpit. On the Parameter Check screen, the following tab pages are available: Deviations Found Intended Deviations Check Environment
Deviations Found
On this tap page, the following information about parameters that deviate from the SAP standard recommendation is displayed: Parameter type Parameter name Member number to which the parameter values apply to Recommended parameter value Current parameter value Deferred parameter value, that is, its value after the database has been restarted
Intended Deviations
On this tab page, the same information as on the Deviations Found tab page is displayed plus the following: User that declared the deviation as intended Time stamp when it was declared as intended Comment Note To display the comment in full length, click the relevant text in the Comment column. End of the note.
September 2011
191
7 Configuration
Check Environment
On this tab page, the following system attributes and attribute values that are relevant for the parameter check are displayed in the System Characteristics area: Attribute Description Indicates if the database is using the database partitioning feature (DPF) The following values are possible: NO DPF In case of a single-partition system YES In case of more than one partition BI Indicates whether the database system has BW content Indicates if the database system is using STMM The following values are possible: NO STMM is not enabled for all partitions. STMM YES STMM is enabled for all partitions. MIXED STMM is only enabled for some partitions. OS BIT Indicates the operating system platform Indicates if the database instances are 32 bit or 64 bit instances
In the Check Configuration area, you are also provided with the number of the SAP parameter note of the underlying DB2 release. You can download the latest version of this SAP Note by choosing the Download SAP Note pushbutton.
Maintaining Parameters
On the Parameter Check screen, you can perform the following actions: Declare parameter deviations as intended Reverse the declared as intended status Correct parameter deviations
192
September 2011
7 Configuration
September 2011
193
7 Configuration
Note In a database environment with multiple database partitions, the deviations of database configuration parameters and registry variables can occur on several partitions. When choosing the Declare as Not Intended menu button, you have to specify if this operation applies to a single or to all partitions where this parameter deviation was found. End of the note. If you want to reverse the status of all deviations on the Intended Deviations tab page to not intended, you can clear the complete list by choosing the Clear List of Intended Deviations pushbutton. As a result, the deviations appear again on the Deviations Found tab page.
194
September 2011
7 Configuration
September 2011
195
7 Configuration
Column DDL_BP_STMM DDL_TBSP DDL_WLM DDL_THRES HADR Object Name Old Value
Description
Name of the object the event relates to If available, the old state or value of the configuration object If available, the old flag of the configuration change Possible values are:
Old Flag
New Value
If available, the new state or value of the configuration object If available, the new flag of the configuration change Possible values are:
New Flag
Invoked by
To display details about a configuration history entry, select the relevant line in the list. The Configuration History Details area appears displaying the full log record.
196
September 2011
7 Configuration
To display details about a buffer pool, choose a buffer pool from the list. The following information is displayed in the Buffer Pool Details area: Technical Settings Database Partition Groups Tablespaces
Technical Settings
The Technical Settings tab page displays the following technical attributes: Field Size Default (BUFFPAGE) Automatic The size of the buffer pool has not been set at creation time, and the default size as defined in the database configuration parameter BUFFPAGE is used. The size of the buffer pool is managed by STMM. Description
September 2011
197
7 Configuration
The size of the buffer pool is set to a fixed value. Fixed Standard Size This means that this size is set for all partitions except for those with exception entries.
I/O Units Page Size Block Size Number of Block Pages Size of one page in the buffer pool Size of one block for blocked I/O in pages (only displayed if you are using DB2 UDB Version 8) Number of pages that are reserved for block I/O usage (only displayed if you are using DB2 UDB Version 8)
For all partitions of the selected buffer pool, the following information is additionally displayed: Column Partition Buffer Pool Size Exception Database partition number Size of the buffer pool in pages Indicates whether an exception entry has been defined for this partition that specifies a size different from the default size in the Fixed Standard Size field Description
Tablespaces
The Tablespaces tab page displays a list of all tablespaces that use this buffer pool as well as the page size of these tablespaces.
198
September 2011
7 Configuration
3. If you want the changes take effect immediately, select the Change Immediately checkbox. Otherwise, the changes will take effect after the next restart of the database. 4. To confirm your entries, choose the Execute pushbutton.
September 2011
199
7 Configuration
Note By default, all available partitions that relate to the new buffer pool are listed. To reduce this list, you can choose the On Selected Database Partition Groups radio button in the Database Partition Groups area, and select single database partition groups. End of the note. 2. Specify a name for the new buffer pool. Recommendation We recommend that you use uppercase letters for the buffer pool name. Using lowercase letters or special characters makes it less comfortable to access the selected database partition group with the DB2 command line processor. End of the recommendation. 3. In the Size and I/O Units areas, enter the relevant technical settings, such as the page size. 4. If the buffer pool is specified for a set of database partition groups only and not yet for all database partition groups, you can add database partition groups by selecting them in the Database Partition Groups list. 5. If you want the new buffer pool to be created immediately, select the Create Immediately checkbox. Otherwise, the newly created buffer pool is created after the next restart of the database. 6. To confirm your entries, choose the Execute pushbutton.
SQL Statements
In the Change Buffer Pool and Add Buffer Pool dialog boxes, the generated SQL statement(s) to be executed are displayed. This area is automatically filled and refreshed as soon as any changes are entered correctly.
200
September 2011
7 Configuration
Overview
This tab page provides an overview of all configured workloads, service superclasses, and their subordinate service classes on the database system. The hierarchy of service classes is displayed using the inheritance arrows that are known from UML diagrams.
September 2011
201
7 Configuration
Workloads
This tab page provides details of all existing workloads as follows: Column Position Workload Name Description Evaluation order that is used for choosing a workload Name of the workload Workload state The following states are possible: Green Enabled The workload is enabled. Red The workload is disabled. DB Access Allowed Service Class Parent Service Class Determines whether or not units of work (UOWs) that are associated with the workload are rejected Name of the service subclass to which an UOW that is associated with this workload is assigned Name of the service super class to which an UOW that is associated with this workload is assigned
To display additional information, you can select a workload from the list. The following tab pages are displayed in the content detail area: Workload Details Displays details about the workload from the system catalog as well as details about the configuration for collecting statistics on the workload Connection Attributes Displays all the attributes that associate an incoming activity with the selected workload For an activity to be associated with a certain workload, all the connection attributes of the incoming activity must match the definition of the workload (Boolean AND). If a single attribute type is specified more than once, this indicates that only one of them must match the activity (Boolean OR).
202
September 2011
7 Configuration
Service Classes
This tab page provides details of existing service classes as follows: Column Service Class Name of the service class State of the service class The following states are possible: Green Enabled The service class is enabled. Red The service class is disabled. Agent Priority Graphical display of the thread priority of the agents in the service class Thread priority of the agents in the service class in relation to the normal priority of DB2 threads The value DEFAULT indicates that the thread priority was inherited from the parent service class. Graphical display of the agents prefetch priority in the service class Prefetch priority of the agents in the service class Graphical display of the buffer pool priority in the service class (only applies to DB2 V9.7) Buffer pool priority of the service class (only applies to DB2 V9.7) Description
Prefetch Priority Prefetch Priority (Remarks) Buffer Pool Priority Buffer Pool Priority (Remarks)
If you select a service class from the list, the Service Class Details tab page appears in the content detail area displaying further details of the service class.
September 2011
203
7 Configuration
Recommendation We recommend that you choose Collect Base Aggregate Activity Data in the Aggregate Activity Data field for all service classes. End of the recommendation. 3. Save your entries. For information about possible data collection settings, see ALTER SERVICE CLASS statement in the IBM DB2 Information Center at: http://publib.boulder.ibm.com/infocenter/db2luw/v9r7/topic/com.ibm.db2.luw.sql.ref.doc/doc/r0 050551.html
204
September 2011
7 Configuration
September 2011
205
7 Configuration
206
September 2011
7 Configuration
Partitions
The Partitions tab page displays the following information: Column Partition Number of the partition Current status of the partition The following values are possible: Status Status information not available Partition not in partitioning map; containers not yet created Partition not in partitioning map; containers created Description
September 2011
207
7 Configuration
Column
Description Partition in partitioning map; containers created Partition will be dropped after next redistribution
Buffer Pools
The Buffer Pools tab page displays a list of all buffer pools that have been defined for the selected database partition group as follows: Column Buffer Pool Name Total space in KB If you create tablespaces, this information is not displayed. Size of the buffer pool in KB Buffer Pool Size (KB) Value 1 indicates that the default buffer pool size parameter from the database configuration is used (parameter BUFFPAGE). Size of one buffer pool page in bytes Description
Page Size
You can display more details about a buffer pool by selecting it from the list. The detail information is displayed in the Buffer Pool Details area below the Database Partition Group Details area.
Tablespaces
The Tablespaces tab page displays a list of all tablespaces that have been defined for the selected database partition group including their page size in bytes. You can display more details about a tablespace of a buffer pool by selecting the tablespace from the list. The detail information is displayed in the Tablespace Details area that appears below the Database Partition Group Details area.
208
September 2011
7 Configuration
September 2011
209
7 Configuration
SQL Statements
In the Change Database Partition Group and Add Database Partition Group dialog boxes, the generated SQL statement(s) to be executed are displayed. This area is automatically filled and refreshed whenever changes are correctly entered.
210
September 2011
7 Configuration
Active
September 2011
211
7 Configuration
Column A
Description
RUNSTATS may be run by DBA Planning Calendar jobs. This information is only relevant if RUNSTATS is performed by DBA Planning Calendar jobs and not by DB2 automatic RUNSTATS. Profile RUNSTATS Date RUNSTATS Time Indicates whether a RUNSTATS profile was set for the table Date of the last RUNSTATS in the system catalog table
Time of the last RUNSTATS in the system catalog table Number of rows as calculated by the last RUNSTATS
Cardinality (1 indicates that there are no statistics available.) For more details about a table, choose a table from the list. The Single Table Analysis screen appears providing detailed information about the selected table. For more information, see Analyzing and Maintaining a Single Table [page 154].
212
September 2011
7 Configuration
Procedure
1. Call the DBA Cockpit and choose Configuration Database tab page of the DBA Cockpit. Automatic Maintenance on the
The Automatic Maintenance screen appears displaying the following tab pages: o o o o General Automatic Backup Automatic RUNSTATS Automatic REORG
Each of these tab pages provides detailed information about the currently selected settings for each of the functions as well as general information applying to all functions. These tab pages are described in more detail later in this document. 2. To configure the automatic maintenance functions, choose the Configure pushbutton. The Configure Automatic Maintenance Functions wizard appears guiding you through the following configuration steps: Step Introduction Action Provides an overview of the available configuration steps Note General To be able to use the automatic maintenance functions, you must select the checkbox Automatic maintenance is switched on.
September 2011
213
7 Configuration
Action
Specify the maintenance windows during which automatic maintenance is performed by DB2: Online Maintenance Window Time frame with only low activity on the database Offline Maintenance Window Time frame with no activity on the database
Select the required maintenance task, for example, Automatic REORG, for one of the maintenance windows. DB2 then decides if any action is required and automatically triggers the correct action.
Note Since the tab pages for specifying the online and the offline maintenance windows are identical, they are only described once. End of the note. Specify the following parameters: o Online Window / Offline Window o o o Caution The definition of all time-related parameters is combined by AND. Therefore, a valid maintenance window must meet all definitions. End of the caution. Specify the following: o Backup o o Start Conditions Backup Media Backup Operation Mode Time Day of Week Day of Month Month of Year
214
September 2011
7 Configuration
Specify the following: o o RUNSTATS Note You can only specify a table filter if you select Custom as Configuration Profile. End of the note. Specify the following: o o o REORG Configuration Profile Index Reorganization Mode Parameters such as the Maximum Table Size, Reclaim Size for MDC Tables, or whether you want to keep the compression dictionary. Configuration Profile Table Filter
Note You can only specify the index reorganization mode and other parameters if you select Custom as Configuration Profile. End of the note. Summarizes all actions to be performed Summary To apply your configuration entries and to exit the wizard, choose the Execute pushbutton.
September 2011
215
7 Configuration
Online maintenance window is enabled / Offline maintenance windows is enabled Time of Automatic Maintenance
Time
Day of Week
216
September 2011
7 Configuration
Day of Month
Days of months during which the automatic maintenance function is to be performed You can specify days or day ranges that are separated by comma, for example, 1, 710. Months of the year during which the automatic maintenance function is to be performed
To change the settings, choose the Configure pushbutton. For more information, see Configuring Automatic Maintenance Functions [page 213]. End of the note.
Specifies the type of backup (online or offline) Backup Operation Type When an online backup is performed, you are still able to access the database during the online maintenance window. When an offline backup is performed within the offline maintenance window, you cannot access the database. Specifies the priority of the automatic backup over the other automatic maintenance features such as Automatic RUNSTATS or Automatic REORG Priority Note 1 means highest priority. End of the note.
September 2011
217
7 Configuration
Starting Conditions Backups are created more frequently. Therefore, less time is required to recover the database. Optimize for Database Recoverability The following limits apply: Maximum time between backups: 1 day Maximum log space used between backups: 10 MB Indicates the balance between the number of backups and the time for recovery. Balance Between Recoverability and Performance The following limits apply: Maximum time between backups: 7 days Maximum log space used between backups: 25 MB Fewer backups are created. Therefore, more time to recover the database is required. Optimize for Database Performance The following limits apply: Maximum time between backups: 30 days Maximum log space used between backups: 50 MB If you choose this option, you can customize the following parameters: Number of full backups is less than Minimum number of backups Customize Time since last full backup exceeds <value> days Maximum time between backups Log space used since last full backup is more than <value> MB Maximum log space between two backups Number of full backups is less than Time since last full backup exceeds Log space used since last full backup is more than If the number of backups is less than the specified value, a backup is created. If the time since the last backup exceeds the specified value, a backup is created. If the log space exceeds the specified value, a backup is created.
218
September 2011
7 Configuration
Backup Media The backup is created in the specified file systems. File System If you choose File System, you also have to specify the File Paths where the backup is to be created. The backup is created on tape. Tape Device If you choose Tape Device, you also have to specify the Number of Parallel Sessions. The backup is created and stored in IBM Tivoli Storage Manager (TSM). If you choose TSM, you also have to specify the File Paths. The backup is created using the XBSA API for storing the data. XBSA If you choose XBSA, you also have to specify the Number of Parallel Sessions. The backup is created and data is stored using a vendor library. Vendor Library If you choose Vendor Library, you also have to specify the Location (that is, a path and file name of the library) and the Options.
TSM
Note To change the settings, choose the Configure pushbutton. For more information, see Configuring Automatic Maintenance Functions [page 213]. End of the note.
September 2011
219
7 Configuration
220
September 2011
7 Configuration
Column
Schema
Name of the schema to which the table belongs Indicates whether the table is flagged as VOLATILE in the system catalog
Volatile If the table is flagged as VOLATILE, statistics are not used by the optimizer. Active Profile RUNSTATS Date RUNSTATS Time A value of N indicates whether the table is marked as VOLATILE by SAP Indicates whether a RUNSTATS profile was set for the table Date of the last RUNSTATS in the system catalog table Time of the last RUNSTATS in the system catalog table Number of rows as calculated by the last RUNSTATS Cardinality (1 indicates that no statistics are available.) Note To change the settings, choose the Configure pushbutton. For more information, see Configuring Automatic Maintenance Functions [page 213]. End of the note.
September 2011
221
7 Configuration
On the Automatic REORG tab page, the following information is displayed: Parameter General Automatic REORG is switched Enables or disables the automatic REORG function on Specifies a maintenance window for index reorganization Recommendation Index Reorganization Mode We recommend that you reorganize indexes during the online maintenance window. End of the recommendation. Specifies the priority of the automatic REORG over the other automatic maintenance features, such as Automatic RUNSTATS or Automatic Backup. Priority Note 1 means highest priority. End of the note. Parameters Description Enables the SAP default filter criteria for tables that are to be excluded from automatic REORG That is, all table filters in the policy are disabled and the parameters are changed according to the SAP recommendations. Note A full editor for these filter criteria is not provided. End of the note. If you select this option, a copy of the table or index is created in the temporary tablespace, and the table or index is copied to the original tablespace. Use a System Temporary Tablespace with Compatible Page Size Since temporary tablespaces in SAP systems are SMS tablespaces, the required space for defragmentation will be available after the reorganization. Recommendation We recommend that you use a system temporary tablespace. Description
222
September 2011
7 Configuration
Parameter
Description End of the recommendation. Specifies the tables that are excluded from the automatic REORG because of their size
Recommendation We recommend a maximum table size filter of 1,000,000 KB. End of the recommendation. Specifies whether MDC tables are reorganized and which size limit is used.
Note This option is only available as of DB2 V9.7. End of the note. Specifies whether you want to keep or rebuild the compression dictionary A rebuild of the compression dictionary can lead to a better compression ratio but implicates a longer REORG runtime.
Compression Dictionary Recommendation We recommend that you rebuild the compression dictionary. End of the recommendation. In addition, the following information is displayed for tables that are excluded from an automatic REORG: Column Description The excluded tables are divided into the following categories: Tables Excluded by Policy Within the policy, there are some filter criteria for tables to be excluded from the automatic REORG. If you expand this node, the excluded tables are displayed. Tables Excluded by Size If you expand this node, the tables with a size larger than the threshold are displayed. Determining the table sizes online is much too expensive. To obtain the sizes of the tables, you have to schedule the job REORGCHK for all Tables in the DBA Planning Calendar.
September 2011
223
7 Configuration
Description Name of database schema to which the table belongs Table to be reorganized Indexes to be reorganized
Table Size (KB) Size of the table in KB Note To change the settings, choose the Configure pushbutton. For more information, see Configuring Automatic Maintenance Functions [page 213]. End of the note.
224
September 2011
7 Configuration
Column
Description Is there a related data tablespace? Does the data tablespace exist in the database? Is there a related tablespace for indexes? Does the index tablespace exist in the database? Does the name of the tablespace comply with the naming conventions for the customer namespace? Is there a description for the data class?
Name of the data class known to the ABAB Dictionary Name of the tablespace where table data is stored
Name of the tablespace where table indexes are stored Number of tables within the related data tablespace
No. of Indexes Number of indexes within the related index tablespace Category Description Category of the data class Description of the data class
In addition, you can maintain data classes, that is, change, add, or delete them as described in Maintaining Data Classes.
September 2011
225
7 Configuration
226
September 2011
7 Configuration
September 2011
227
7 Configuration
Administrative Task Scheduler, you can migrate the tasks to the DB2 Administrative Task Scheduler using the Migrate to DB2 Admin Task Scheduler link. In the content area, the following information is displayed: Column Name of the data collector Data Collector Typically, the name refers to the data that is collected, for example, the data collector Tables stores table-related metrics in the history tables. Type of data collector The following collector types are possible: SAP Data Collection and Preparation (STP) These data collectors have their own mechanism to calculate the delta of the different snapshots. Type DB2 Event Monitor and Preparation (STP) A DB2 event monitor is used to store the data in tables. Cleanup of Message Logs (STP) The Framework data collector is not a typical data collector type, but a cleanup job for the messages generated by the above mentioned data collector types. Indicates the setup status The following values are possible: Green: The data collector is set up as intended. Yellow: Setup The data collector is not set up properly. We recommend that you repair the data collector. Red: The data collector is not set up. Set up the data collector according to the assigned template. Gray: The data collector is not set up and the assigned template also marks this data collector to be deactivated. Collection Interval Schedule of the data collector Description
228
September 2011
7 Configuration
Description
The following values are possible: Green: The DB2 event monitor is set up as intended. Yellow: Event Monitor Active The DB2 event monitor is not set up properly. We recommend that you repair the data collector. Red: The DB2 event monitor is not set up. Set up the data collector according to the assigned template. Gray: No DB2 event monitor is used by the data collector. Available Days Days Kept in History Size on Disk (MB) Errors Availability of the collected data for the given number of days as of now Number of days of the sliding history window The data in the history is deleted after this number of days. Space consumption in the local history of the data that is stored by the data collector Number of errors that the data collector has captured and that the scheduler has recorded Number of warnings that the data collector has captured and that the scheduler has recorded
Warnings
September 2011
229
7 Configuration
Depending on your setup status, you can proceed with the following maintenance actions: Action Setup Status Steps to Be Performed On the Configuration tab page, choose the Set Up pushbutton. Setting up a Gray or red as the data single history data collector is not set up yet collector Note You must set up the Framework collector before all other data collectors. End of the note. Yellow as the data Repairing a single collector is set up but history data should be updated or collector needs to be repaired
1. On the Configuration tab page, choose the Change Settings pushbutton and enter your changes. 2. To apply your changes, choose the Apply Changes pushbutton. If you want to discard your changes, choose the Reset Changes pushbutton. On the Configuration tab page, choose the Prune pushbutton. By doing so, the collected data and the generated messages of the data collector are removed. On the Configuration tab page, choose the Drop pushbutton. On the Data Collector Log tab page, you can choose between the following pushbuttons:
Removing data of a single history Yellow or green data collector Dropping a single history data Yellow or green collector
230
September 2011
7 Configuration
September 2011
231
7 Configuration
The following figure provides an overview of the workloads and service classes that are created:
Workloads
SAP_ABAPDIA_WL
Service Subclasses
SAP_ABAPDIA_SSC
Service Superclasses
SAP_ABAPBTC_WL
SAP_ABAPBTC_SSC
SAP_ABAPSPO_WL
SAP_ABAPSPO_SSC
SAP_ABAPUPD_WL
SAP_ABAPUPD_SSC
SAP_SC
SAP_ABAPUPD2_WL
SAP_ABAPUPD2_SSC
SAP_JAVA_WL
SAP_JAVA_SSC
SAPTOOLS_WL
SAPTOOLS_SSC
SAP_DIAGNOSTIC_WL
SAP_DIAGNOSTIC_SSC
DB2_ATS_WL
DB2_ATS_SSC
DB2_SC
Overview of Workloads and Service Classes The SAP_DIAGNOSTIC_WL workload as well as the SAP_DIAGNOSTIC_SSC service class are created for SAP-internal use. Currently, no work is executed in this service class. The SAP_SAPTOOLS_WL workload as well as the SAP_SAPTOOLS_SSC service class are created to separate work that originates from SAP monitoring tools. You can monitor the SAP_SAPTOOLS_SSC service class to determine the amount of resources that are consumed by SAP monitoring tools such as the DBA Cockpit or the Performance Warehouse. The DB2_ATS_WL workload as well as the DB2_ATS_SSC service class are created to separate work that is executed by the DB2 task scheduler. The DCF uses the DB2 task scheduler to execute the data collectors. Since in typical SAP systems no other component uses this mechanism, the workload that is measured on this service class is caused by the DCF of the DBA Cockpit. The service superclass SAP_SC is only created for structural reasons. No work is executed in service superclasses. In addition to the default workloads and service classes that are created in SAP environments, you can set up additional workloads and service classes as described in Workload Management: SAP WLM Setup Status [page 205].
232
September 2011
7 Configuration
Procedure
1. Call the DBA Cockpit and choose Configuration Database tab page of the DBA Cockpit. Monitoring Settings on the
2. On the Monitoring Settings screen, choose the Change pushbutton. 3. Enter your changes in retention time of the statistics in the relevant fields under Database and Tablespace Size, Table and Index Size, and Database Performance. 4. If you want to switch the db2diag.log file to automatically restrict its size to a manageable value, select the Switch Weekly checkbox. This means that the db2diag.log file is saved under a new name with a time stamp and a new db2diag.log file is created. 5. If you want to collect history data on a dedicated background server, specify a server in the Server for Data Collection field by choosing a name from the dropdown list. 6. Save your changes.
September 2011
233
8 Jobs
8 Jobs
The following sections provide information about the main task areas that are available under Jobs on the Database tab page of the DBA Cockpit: DBA Planning Calendar DBA Log Back-End Configuration SQL Script Maintenance Scheduled DB2 Tasks
Integration
The DBA Planning Calendar is part of the DBA Cockpit. You can start the DBA Planning Calendar by calling the DBA Cockpit and choosing Jobs DBA Planning Calendar on the Database tab page of the DBA Cockpit.
Features
The initial screen of the DBA Planning Calendar consists of a Selection area and the calendar. In the following, both are described in more detail.
234
September 2011
8 Jobs
DB Collectors These are actions that are automatically selected by the system to collect data, for example, on performance or history and that are only available for RFC-monitored systems.
Select the week to be displayed The default is the current week. To navigate to another week, use the F4 help. Select a factory calendar Specifying a factory calendar only has an impact on the calendar display. Holidays are the same color as weekend days. It does not result in any restrictions on planned actions.
The Calendar
The calendar displays all scheduled jobs. The status of each job is indicated using different colors. To display the meaning of each color, choose the Legend pushbutton. The following information is displayed: Color Light blue Dark blue Green Status Planned Description The action has not yet started. Action
Running Finished successfully Finished with warning Finished with error No longer available Scheduling failed
The action has run successfully. The action has finished with a warning. An error has occurred and the action was interrupted. No more information is available. Scheduling failed, that is, there is no status available and the action is overdue.
Check the job log for details. Check the job log for details and reschedule the action.
Yellow
Note By default, the DBA Planning Calendar appears in display mode. To switch to editing mode, choose the Change pushbutton in the global toolbar. End of the note.
September 2011
235
8 Jobs
More Information
Setting Up and Working with the DBA Planning Calendar Scheduling Typical DBA Actions
More Information
Scheduling Typical DBA Actions [page 244]
236
September 2011
8 Jobs
If the DCF is not enabled and the SAP history collection is used instead, the following jobs are involved: Collection of database performance history data done every two hours starting at 00:00 Monitoring of database and database manager configuration changes done daily at 8.00 am, 01.00 pm, and 7.00 pm Collection of database and tablespace history data done daily at 7:00 am and 8.00 pm Collection of space history data for tables and indexes done weekly on Sundays at 12.00 pm Caution Some of the available actions have an impact on database performance and availability. Check the start and end times of scheduled actions to make sure that they do not interfere with each other and that they do not impact other system activities. You cannot perform all required DBA actions from the DBA Planning Calendar or the DBA Cockpit. For more information about actions that you must perform with the SAP system down, such as an offline database backup, see the SAP Database Administration Guide for your database version. End of the caution. Optionally, you can configure the back end of the DBA Planning Calendar to be able to control the execution of these background jobs. For more information, see Back-End Configuration [page 258].
Prerequisites
Before you start using the DBA Planning Calendar, you have to check the following: SAP system authorizations Check that you have authorization for DBA and background job scheduling, which is provided by profiles S_RZL_ADMIN and S_BTCH_ALL. Check that external programs are able to run on the database server so that actions affecting the database can be performed from other application servers. Database system authorizations Check that actions requiring authorization to connect to the database are permitted. Some actions, such as online backup, require higher levels of authorization. Before scheduling database backups, see the information about backing up the database in the SAP Database Administration Guide for your database. Hardware and backup media Check that you have enough hardware (such as tape drives) and backup media (such as tapes) for the backup strategy you intend to use.
September 2011
237
8 Jobs
You make sure that the media are initialized and ready so that the operator does not have to interrupt scheduled backup runs. For example, check that tapes are already in the specified tape drive or tape changer. You might require different tape drives for database backups and log backups.
Procedure
1. On the Database tab page of the DBA Cockpit, choose Calendar . Note By default, the DBA Planning Calendar appears in display mode. To switch to editing mode, choose the Change pushbutton in the global toolbar. End of the note. 2. Choose the Pattern Setup pushbutton. The Setup Planning Pattern wizard appears guiding you through the procedure of setting up a pattern of recurring actions that cover your regular DBA needs. Note Depending on your database version and configuration, a set of recommended actions is available and some of these actions might be preselected by default. You may change this selection set, for example, if you want to use TSM for backup and archiving instead of backup and archiving to devices. End of the note. 3. To enter the pattern into the DBA Planning Calendar, choose the Execute pushbutton on the Summary screen. 4. Exit the wizard. Caution When you start using the DBA Calendar in production operation, you must check daily that scheduled actions have been executed correctly. End of the caution. Jobs DBA Planning
238
September 2011
8 Jobs
The Schedule an Action wizard appears. 2. In the Job Selection step, do the following: 1. Choose the job that you want to schedule from the Action dropdown list. 2. Specify one of the following options: Option Start Immediately Start on Description The job starts immediately in the background after you have completed the required steps in the wizard. Specifies the date and time when the action is to start
Start as Recurring Activates the Recurrence step where you can further specify a recurrence Action pattern for this job as described later on in this section 3. In the Parameter step, change, or enter the basic parameters for the action. 4. Optional: If you previously selected the Start as Recurring Action option in the Job Selection step, you now have to enter a recurrence pattern in the Recurrence step. The following table describes the parameters to be specified in more detail: Parameter Description Interval for the action in weeks, days, or hours Recurrence Pattern Depending on the selected recurrence pattern, you need to specify the pattern in more detail, that is, the days of the week for weekly periods and the hours of the day for a daily period. The action is repeated at the interval that you enter. If you select Once only, the action is executed only once. Recurrence Range Range of time where the action recurs, that is, for a specific time interval or for a limited number of occurrences
September 2011
239
8 Jobs
Caution The system warns you if there is a conflict with an existing action, but it does not prevent you from inserting the new action. You must decide whether the actions might conflict in terms of database access or performance. The system does not check for conflicts between actions with identical start times but checks for actions within a range of approximately 30 minutes. End of the caution. 5. On the Summary screen, review the specified parameters and schedule the job by choosing the Execute pushbutton. 6. Exit the Schedule an Action wizard.
Result
The schedule of the DBA Planning Calendar is updated.
Prerequisites
The action to be changed must be in the state Planned (that is, not already executed). Note If an action has already been executed, you can only display it. For more information, see Displaying Scheduled Actions [page 242]. End of the note.
Procedure
1. In the DBA Planning Calendar, click a calendar cell that contains an action. The Action Description detail area appears displaying the action parameters and, if available, the recurrence pattern. 2. From the Change menu button, choose one of the following options: o If you want to edit the currently selected occurrence only, choose Selected Occurrence.
240
September 2011
8 Jobs
If you want to change all occurrences of a recurring job, choose All Occurrences.
The Change Scheduled Action dialog box appears displaying the action parameters and, if available, the recurrence patterns. 3. Apply and save your changes.
Prerequisites
The action to be deleted must be in the state Planned (that is, not already executed).. Note If an action has already been executed, you can only display it. For more information, see Displaying Scheduled Actions [page 242]. End of the note.
Procedure
1. In the DBA Planning Calendar, click a calendar cell that contains the action to be deleted. The Action Description detail area appears displaying the action parameters and, if available, the recurrence pattern. 2. From the Delete menu button, choose one of the following options: o o If you want to delete the currently selected occurrence only, choose Selected Occurrence. If you want to delete all occurrences of a recurring job, choose All Occurrences.
Procedure
1. Click the action that you want to re-execute. The Action Description detail area of the action appears where you can check the action parameters. 2. Choose the Re-Execute pushbutton.
Result
The action is rescheduled starting at the current time.
September 2011
241
8 Jobs
Procedure
To display a scheduled action, click a calendar cell that contains an action. The Action Description detail area appears displaying scheduling information or the return status of a finished action. The following tab pages are displayed: Action Parameters Displays the parameters of the action Recurrence Displays the recurrence patterns and only appears if the action is part of a recurring action Job Log The background processing job log generated by the action is displayed under Job Log. All messages that have been written by the background job are also displayed. To display long texts, if any are available, double-click a message. Program Log Some actions write log files onto the database server. If such a program log exists, it is displayed on this tab page.
242
September 2011
8 Jobs
8.1.1.7 Troubleshooting
Since any action scheduled in the DBA Planning Calendar can fail, you must at least check the more critical actions such as database backups. The following steps describe how you check whether a job was executed correctly and how to proceed if it failed.
Procedure
1. Check the status of the job in the DBA Calendar. The status of a job is indicated by different colors. If the job finished successfully, the calendar cell is marked green. 2. If the job terminated with a warning, an error, consult the job log. Note If there is no job log available, the background job was probably not started. In this case, proceed as follows: 1. In your SAP system, call transaction SM37 and display the job overview. The jobs related to the planning calendar start with the prefix DBA:. The complete name of such a job in SM37 contains a short description of what the job is about. The following is an example of, for example, a REORG job: DBA:REORG_TABLE_____@095534/4000. 2. To search for information about possible errors that prevented the job from being executed, double-click the respective background job in the overview list. End of the note. 3. Correct the error using the information that you found the job log or in transaction SM37, and reschedule the job in the DBA Planning Calendar. Make sure there are no conflicts with other scheduled actions before you reschedule this job. Note If you want to completely clean up your jobs, choose the Cleanup pushbutton. This deletes all jobs, all scheduling data, and all related protocol records. It also resets the DBA Planning Calendar to its initial state. We recommend that you clean up after an SAP system upgrade or if jobs have become corrupt. End of the note.
September 2011
243
8 Jobs
244
September 2011
8 Jobs
Snapshot Backup You back up the database using the fast copying technology of a storage device. Caution This job is only available as of DB2 V9.5. End of the caution.
Procedure
1. To schedule a database backup in the DBA Planning Calendar, click a calendar cell or choose the Schedule an Action link in your Favorites. The Schedule an Action wizard appears guiding you through the procedure. In the Parameters step, specify the following parameters: Parameter Backup Mode Online Access to the database is not blocked. The users can continue to work normally during the backup. Note This option is no longer supported and is displayed only for upward compatibility reasons. Backup jobs that have this option are automatically performed as online backups with the Include Logs option. End of the note. Backup Type Full The complete database is backed up. Cumulative (that is, incremental) backup image Incremental An incremental backup image is a copy of all database data that has changed since the most recent successful full backup operation. Non-cumulative (that is, delta) backup image Incremental Delta A delta backup image is a copy of all database data that has changed since the most recent successful backup operation. Description
Offline
September 2011
245
8 Jobs
Additional Options Compress The backup is to be compressed. Note Only choose this option if you want to perform an online backup. Include Logs End of the note. Only those log files are included in the backup that are required to get a consistent database. Any further log files are not taken into consideration. Caution The following options are only available for downward-compatibility reasons and we strongly recommend that you do not set them: o o o Number of Buffers Buffer Size Parallelism
End of the caution. 2. Schedule the job by choosing the Execute pushbutton on the Summary screen of the Schedule an Action wizard. 3. Exit the wizard
246
September 2011
8 Jobs
September 2011
247
8 Jobs
More Information
Database Administration Guide: SAP on IBM DB2 for Linux, UNIX, and Windows at: http://service.sap.com/instguidesnw Database-Specific Guides <Your SAP NetWeaver Release> Operations
Procedure
1. Start the Schedule an Action wizard by clicking a calendar cell or by choosing the Schedule an Action link in your Favorites. 2. In the Job Selection step, choose Archive Log Files to Tape from the Action dropdown list. 3. In the Parameter step, specify the required parameters. Note The DB2 tape manager is used to archive log files to tape. Besides the standard parameters (for example, start time, date, number of log files to be saved, and tape label), you can also specify the option of the tape manager to use for archiving log files: o o o Double Store Overwrite Expired Tapes Eject Tape at End of Operation
For more information, see the Database Administration Guide SAP on IBM DB2 for Linux, UNIX, and Windows. End of the note. 4. Schedule the job by choosing the Execute pushbutton on the Summary and exit the wizard.
More Information
Database Administration Guide SAP on IBM DB2 for Linux, UNIX, and Windows at: http://service.sap.com/instguidesnw Database-Specific Guides <Your SAP NetWeaver Release> Operations
248
September 2011
8 Jobs
Procedure
1. Start the Schedule an Action wizard by clicking a calendar cell or by choosing the Schedule an Action link in your Favorites. 2. In the Job Selection step, choose RUNSTATS and REORGCHK for Set of Tables from the Action dropdown list. 3. In the Parameters step, do the following: o o Specify the tables where the statistics are to be updated. You can use wildcards for the table names. Specify how you take statistic profiles into account by choosing one of the following options: Use Customized Settings or Existing Profiles For tables that have statistic profiles, the RUNSTATS is performed using exactly the same settings as specified in the profile. For all other tables, the RUNSTATS is performed as specified in the wizard. Use Customized Settings Ignoring Existing Profiles If you choose this option, the RUNSTATS is performed as specified in the wizard regardless whether or not statistic profiles exist. o Specify the following settings for the RUNSTATS: Field Table Analysis Method Basic Distributed Statistics Basic statistics for the table Distributed statistics for the table Description
September 2011
249
8 Jobs
Description
Selecting this option only freezes already existing old table statistics but it does neither delete nor invalidate them. End of the caution. Percentage of entries to be used for sampling Sampling of [ ] %of entries Caution This field is only active if you are using DB2 UDB for UNIX and Windows Version 8, FixPak 2 or higher. End of the caution. The data to be sampled is selected page by page. System (Page, Sampling) Caution This field is only active if you are using DB2 UDB for UNIX and Windows Version 8, FixPak 2 or higher. End of the caution. The data to be sampled is selected row by row. BERNOULLI (Row Sampling) Caution This field is only active if you are using DB2 UDB for UNIX and Windows Version 8, FixPak 2 or higher. End of the caution. Analyze Key Columns Table statistics are gathered only for key columns of the table. only Index Analysis Method Basic Detailed Statistics Detailed Sampled Statistics Basic statistics for the index are gathered. Detailed statistics for the index are gathered. Detailed statistics for the index using sampling are gathered.
250
September 2011
8 Jobs
No statistics for the index are gathered. Caution None Selecting this option only freezes already existing old index statistics but it does neither delete nor invalidate them. End of the caution. o If you have specified more than one table, you can choose the degree of parallelism by setting the value for Number of Parallel Jobs higher than one. In this case, the RUNSTATS job can be parallelized if there are enough system resources available (for example, background processes and the appropriate number of processors). Doing so results in additional jobs that are scheduled by the regular DB13 job and which perform the RUNSTATS on tables in parallel. The SAP system makes sure that the number of parallel jobs does not exceed the number of available background processes. Caution Handle the parameter Number of Parallel Jobs with care. Starting more jobs can have a high impact on the overall system performance even though the parameter Number of Parallel Jobs significantly reduces the amount of time for the job execution. End of the caution.
o o
4. Schedule the job by choosing the Execute pushbutton on the Summary screen and exit the wizard.
September 2011
251
8 Jobs
Procedure
1. Start the Schedule an Action wizard by choosing the Schedule an Action link in your Favorites or by clicking a calendar cell. 2. In the Job Selection step, choose REORG and RUNSTATS for Set of Tables from the Action dropdown list. 3. In the Parameters step, specify the following parameters: Parameter Table Schema Table Description Name of the schema to which the table belongs Name of the table Schedules an offline reorganization Optionally, you can also specify the following parameters: o Use Temporary Tablespace If you select this option, a temporary tablespace is used for the reorganization. Note If no temporary tablespace is used for the REORG, it is performed in the tablespace where the table or index resides. You must make sure that there is enough free space in this tablespace (approximately the size of the table or index). If this tablespace already has a high I/O load, we recommend that you use a temporary tablespace for performance reasons. End of the note. o With Long Fields and LOB Data If you select this option, long field and LOB data areas are also reorganized. o Switch Compression on If you select this option, the table and its indexes are enabled for compression before the REORG operation is performed. o Reset Dictionary
Offline
252
September 2011
8 Jobs
Parameter
Description If you choose this option, a compression dictionary is rebuilt if the table is compressed or the Switch Compression on option is set. Otherwise, an existing compression dictionary is removed. o Keep Dictionary If you choose this option, an already existing compression dictionary is kept and not rebuilt. If the table is compressed or the Switch Compression on option is set and a dictionary does not yet exist, the compression dictionary is built. Note The compression-related options are only available for DB2 V9.1 and higher. End of the note.
Schedules an online reorganization of the table Schedules only a reorganization of all indexes Schedules a reorganization that reclaims space for MDC tables
Note This option is only available for DB2 V9.7 and higher. End of the note.
4. Schedule the job by choosing the Execute pushbutton on the Summary screen and exit the wizard.
September 2011
253
8 Jobs
Procedure
1. Start the Schedule an Action wizard by clicking a calendar cell or by choosing the Schedule an Action link in your Favorites. 2. In the Job Selection step, choose REORGCHK for All Tables from the Action dropdown list. 3. In the Parameters step, specify the required parameters as follows: Parameter Description Specifies that the job is called for all tables All Tables By default, this parameter is selected. Specifies that this job is restricted to a set of tables If you choose this option, you also have to specify the Table Schema and Table Name. Selected Tables Note Only choose this option if you require an up-to-date analysis for the selected tables. End of the note. Analyzes the tables and checks for candidates to be compressed With Compression By default, you should not activate this option for the REORGCHK job that Check is scheduled weekly. For performance reasons, only perform compression checks in larger time-frames. Defines the minimum size limit for checking how much space can be saved by compressing the table. Recommendation We recommend that you set this limit to prevent too small tables that do not benefit from row compression from being checked. End of the recommendation. Note You can only specify this value if you have chosen With Compression Check. End of the note. Maximum Runtime Restricts the runtime of this job
254
September 2011
8 Jobs
4. Schedule the job by choosing the Execute pushbutton on the Summary screen and exit the wizard.
Procedure
1. Start the Schedule an Action wizard by clicking a calendar cell or by choosing the Schedule an Action link in your Favorites. 2. In the Job Selection step, choose CLP Script from the Action dropdown list. 3. Specify SQL statements directly as job parameters. Note Alternatively, you can use scripts that have been created before. For more information, see The SQL Script Maintenance [page 259]. End of the note. 4. Save your entries by choosing the Execute pushbutton on the Summary screen. 5. Exit the wizard
Procedure
1. Start the Schedule an Action wizard by clicking a calendar cell or by choosing the Schedule an Action link in your Favorites. 2. In the Job Selection step, choose NLS Cleanup from the Action dropdown list.
September 2011
255
8 Jobs
3. In the Parameter step, specify the following: o o In the NLS Connection field, choose an existing NLS connection. In the InfoProvider field, choose the name of the InfoProvider.
Caution If you do not specify an InfoProvider, the cleanup is performed for all InfoProviders. To run the job for many InfoProviders with similar names, you can use the wildcard character *. End of the caution. 4. Save your entries by choosing the Execute pushbutton on the Summary screen. 5. Exit the wizard.
More Information
For more information about reloading data from the NLS database into the BW system, see the document Enabling an SAP NetWeaver BW to Use IBM DB2 for Linux, UNIX, and Windows as Near-Line Storage at: http://service.sap.com/instguidesnw <Your SAP NetWeaver Release> Installation - SAP NetWeaver Systems Installation
256
September 2011
8 Jobs
In the Selection area, you can specify a time frame for which you want to display data. After you have applied your selection, the following information is displayed: Column Partition Start Date Start Time End Date End Time Runtime Action Return Code Name of the partition Start date of the action Start time of the action End date of the action End time of the action Runtime in HH:MM:SS Description of the action Return code of the action Description
When you access the DBA log for the first time, the log information for the current week is displayed by default. For more information about an action, select an entry from the overview table. The Log Details area appears displaying more details about the log file.
September 2011
257
8 Jobs
Configuration Steps
1. Call the DBA Cockpit and choose Database tab page. Jobs Back End Configuration on the
2. To switch to editing mode, choose the Change pushbutton. 3. In the Selected Scope area, choose the scope of the configuration entry. 4. In the Background Dispatching area, specify the following parameters: Field Description Server where scheduled jobs are executed Background Server If no server is specified, the background job dispatcher dynamically selects the server. Priority of the job Job Priority If no priority is specified, jobs run with default priority (low). Name of the user to execute the job User If no user is specified, the dialog user is used. 5. Save your changes.
258
September 2011
8 Jobs
Integration
The SQL Script Maintenance is part of the DBA Cockpit.
Activities
To access the SQL Script Maintenance, call the DBA Cockpit and choose Script Maintenance on the Database tab page. Jobs SQL
The SQL Script Maintenance screen appears displaying all the scripts located on your local monitoring system. You can perform one of the following actions: Action Display an existing SQL script Edit an existing SQL script Delete an existing SQL script Steps to Be Performed Choose an entry from the list. Detailed information is displayed in the SQL Script Details detail area. 1. Choose a script from the list and choose the Edit pushbutton. 2. Enter and save your changes in the Change Script dialog box. 1. Choose a script from the list and choose the Delete pushbutton. 2. Confirm the deletion. 1. To add a new script, choose the Add pushbutton. Add a new SQL script 2. In the Add Script dialog box, enter a name in the Script Name field and start editing. 3. Save the new script. 1. Choose a script from the list and choose the Execute pushbutton. 2. In the Execute SQL Script dialog box, enter the name of the SAP system where you want the script to be executed in the Execute SQL Script on System field. 3. To confirm your entry, choose the Execute pushbutton again. The CLP Output Tray appears showing an editor in the lower half of the screen where the result is displayed. Note As an alternative to the SQL Script Maintenance function, you can also use the DBA Planning Calendar to execute a script by using the CLP Script job as described in Scheduling Scripts [page 255]. End of the note.
September 2011
259
8 Jobs
In the Selection area, you can specify the time frame to be displayed. By default, the jobs of the current day are displayed. The following information is displayed: Column Name Name of the task Status of the task The following values are possible: RUNNING The task is currently running. COMPLETED Status The task has finished running. NOTRUN An error prevented the scheduler from calling the tasks procedure. UNKNOWN The task started running but an unexpected condition prevented the scheduler from recording the task outcome. This can occur if the system ends abnormally or a power failure happens while the task is running. Invocation Number of invocations so far Starting time of the task Time Started If the status is RUNNING, COMPLETED, or UNKNOWN, this value indicates the time when the task started running. The status is NOTRUN indicates the time that the task should have started. Time when the task finished running Time Finished This value is not set as long as the status is RUNNING. If the status is UNKNOWN, this value is the time when the task scheduler detected that the task was no longer executing and updated the status table accordingly. Description
260
September 2011
8 Jobs
Column
Description
Agent TID Agent thread ID of the application that executed the task Value displayed depends on the status this field. If the status is COMPLETED, this value indicates the SQL code that is returned by the CALL to the procedure. If the status is NOTRUN, this value indicates the SQL code of the error that SQL Code prevented the task from running. Note As long as the status is RUNNING or UNKNOWN, this value is not available. End of the note. SQL State The SQL state that is related to the SQL code Return Code Error Message Return code of the procedure that was called As long as the status is not COMPLETED, this value is not available Full message text corresponding to the SQL Code
September 2011
261
9 Alerts
9 Alerts
The following sections provide information about alerts: Alerts: Database System Monitoring in CCMS Alerts: Configuring Database System Monitoring Alerts: Alert Monitor Alerts: Alert Message Log Alerts: Alert Configuration
Integration
The monitoring functions are fully integrated into the CCMS alert monitoring and architecture.
Features
The following categories of information are monitored: Disk space of the tablespaces and file systems that are required for the database system Performance indicators such as the following: o o Access behavior of database buffers Lock behavior of the application, monitoring of deadlock situations, and lock escalations
Availability of backup and recovery mechanisms: o o Last available backup Availability of the log files that are necessary to achieve the current state of the database from the last available backup
Configuration parameters Availability data of tablespaces and containers Recommendation We recommend that you check the alert monitor daily for information about database system monitoring. End of the recommendation.
262
September 2011
9 Alerts
The Monitoring: Tool Administration screen appears. 5. Choose Parameter. 6. Choose Tool Definitions tool bar). Display Change (or Display Change in the application
7. In the SENDER line in the Parameter value column, enter a valid user for your SAP system.
September 2011
263
9 Alerts
8. In the RECIPIENT line in the Parameter value column, enter a valid user for your SAP system who will be notified in the event of an alert. 9. Save your changes.
264
September 2011
9 Alerts
September 2011
265
9 Alerts
266
September 2011
9 Alerts
The checked parameters are displayed in the following colors depending on the type of message: Message Type Information Warning Error Green Yellow Red Color
If a check resulted in a warning or error, a short text is additionally displayed next to the parameter in the Open alerts view.
September 2011
267
9 Alerts
Thresholds
This tab page displays the threshold statuses of an alert configuration as follows: Error Condition If the error is not corrected, normal operation is endangered. Warning Condition Limited operation, for example, with reduced performance Normal State Condition Each condition is described by a relational operator, a comparison value, and the unit of measurement for this value.
268
September 2011
9 Alerts
General (RZ21)
This tab page displays the scheduling data from the basic alert monitor configuration for completeness reasons. You cannot change the values on this tab page. You can make changes using the general maintenance function (transaction RZ21) in the alert monitor.
Administration
This tab page displays the user, date, and time of the last changes.
September 2011
269
10 Diagnostics
10 Diagnostics
The following sections provide information about the main task areas under Diagnostics on the Database tab page of the DBA Cockpit: Audit Log Lock-Wait Events Missing Tables and Indexes Message History Dump Directory CLP Commands DB2 Logs Trace Status Cumulative SQL Trace
From System System from where the action was performed In the Selection area, you can specify a time frame for which you want to display data. By default, the audit logs of the current week are displayed.
270
September 2011
10 Diagnostics
Event Occurrence Time when the lock event first occurred Time when the lock event was solved Event Resolution Note For events that are still existing, the status displayed is Still Existing. End of the note. Involved Agents Max. Wait Time Lock Objects Number of agents involved in the lock event Max. lock-wait time of any of the involved agents Database objects involved in the lock conflict
September 2011
271
10 Diagnostics
Activities
You can display the lock chain - or in case of a deadlock a log circle - of an event graphically by choosing a line from the overview table. To display details about the agents involved, select the appropriate agent in the graphical display. The Details of Agent <agent numer> detail area appears providing general information about the application the lock-waiting agent is working for as well as about the application the lock-holding agent is working for. In addition, information about the activity that is currently being executed or that was executed last by each agent is provided, such as, the SQL statement text and general information about the SQL statement (for example, the compilation environment). To show the current access plan of the statement, you can choose the EXPLAIN pushbutton. For more information, see The EXPLAIN Function [page 43]. To display the ABAP source code from where the SQL statement was executed, you can choose the Show Source pushbutton. Note The pushbutton is only available if the following applies: The SAP system on which the currently monitored database is running is an ABAP system. The SAP system has the minimum Support Package level as described in SAP Note 1493490. End of the note. Note If the lock event monitor has captured past activities or nested activities for a certain agent, the Activity History table appears providing information about the agent ID; its nesting level and the SQL statement executed by each agent. You can display detailed information for each activity by choosing a line in the Activity History table. The information is displayed in the Details of Activity <activity number> detail area. End of the note.
272
September 2011
10 Diagnostics
Objects missing from Objects that are defined in the ABAP Dictionary, but not found in the the database database Unknown objects in Objects that are found in the database, but not defined in the ABAP the ABAP Dictionary Dictionary Inconsistent objects Results of the detailed comparison of the ABAP Dictionary and the database are displayed here Different checks are performed here: It is checked whether the primary index of tables defined in the ABAP Dictionary was created uniquely on the database. Other checks Objects in the SAP system tables are checked, which cannot be described at all or which cannot be completely described in the ABAP Dictionary for technical reasons. If inconsistencies for these objects are detected, they are also displayed here. In general, additional information about the type of inconsistency is provided. Optional indexes Mismatch between ABAP Dictionary and database regarding secondary indexes
If the database structure has been changed since the last consistency check, choose the Refresh pushbutton. Recommendation To ensure consistency between the ABAP Dictionary and the database, the consistency check should be performed once a month or whenever the database structure has changed. End of the recommendation.
September 2011
273
10 Diagnostics
274
September 2011
10 Diagnostics
September 2011
275
10 Diagnostics
How to maintain the blacklist is described later in this section. Choose one of the following options in the Display Type field: Show All Messages No grouping is applied, but all messages are displayed. By default, this option is preselected. Group by Message Code In the result list, all occurring messages are grouped by message code. Group by Message Text In the result list, all occurring messages are grouped by message text. Note If you choose one of the grouping options, only the first message of the chosen grouping type is displayed in detail. The Count column indicates the total number of messages of this grouping type. End of the note. After you have applied your selections, the following information is displayed: Column Time of Collection Description Time when the message was created Time when the message was created Note Time Stamp of This column replaces the Time of Collection column if you have chosen one First Message of the grouping options as display type. Only the first message is displayed in detail whereas the other messages are only added up and given as a total number. End of the note. Identifier for the database member from which the data was retrieved Member Note This column is only visible in a DPF system. End of the note. Total number of messages matching the Display Type criteria Count Note This column is only visible if you have chosen one of the grouping options as display type.
276
September 2011
10 Diagnostics
Description
Message type that is related to the message identifier Message Type Example ADM is used for administration or notification log messages. End of the example. Message number of the associated message Message Number Example 6102 is the numerical part of the ADM6102I message code. End of the example. Indicates the severity level of the message The following values are possible: C Indicates a critical situation E Indicates that an error occurred Level I Displayed message is only informational S Indicates a severe situation, for example, deadlocks have been detected W Indicates that the message contains a warning
September 2011
277
10 Diagnostics
Qualifies the impact of the message from a DBAs perspective The following values are possible: CRITICAL Impact IMMEDIATE NONE POTENTIAL UNLIKELY Message Text Short description of the message
278
September 2011
10 Diagnostics
To avoid unnecessary system workload, you can restrict the amount of data that is processed by specifying the following in the Selection area: A Log Facility A Record Type The minimum impact level (Impact) A time range (Messages From / To)
September 2011
279
10 Diagnostics
If you apply your selection, information according to the specified values is displayed including the appropriate DB2 component. To display additional details about a log entry, click the relevant entry in the list.
DBSL Trace Minimum Time If provided, only SQL statements with execution times higher Limit than this time limit are traced.
280
September 2011
10 Diagnostics
Cumulative Trace First Trace Entry Displays the start time of this trace if trace information already exists Last Trace Entry Displays the end time of this trace if trace information already exists Number of Entries Deadlock Trace Detection Interval Only SQL statements running longer than this time are recorded for deadlock detection. Displays the number of entries in this trace if trace information already exists
For each trace, a pushbutton shows whether the trace is active or switched off. To change the information to be displayed on the Trace Status screen, choose the Change pushbutton. You can then perform one or more of the following actions: Choose if you want the trace status to be displayed on the connected application server or on all application servers by using the relevant radio button. Activate or deactivate the trace function by choosing the Trace Off or Trace On pushbutton. Change the trace level. Maintain the trace parameters in the DBSL Trace Attributes area.
September 2011
281
10 Diagnostics
SQL Statement
If you want to display more detailed information, choose an entry from the list. The Cumulative SQL Trace - Details area appears providing information about the optimization level, the query degree as well as the first and last statement usage. In addition, the following tab pages are displayed: Statement Information Displays the complete SQL statement, the application server where the statement was executed, and a list of all ABAP reports in which the statement can be found Time Histograms Displays the distribution times of the selected SQL statement If you want to display the access plan for the execution of a single statement, select the relevant entry and choose the Explain pushbutton. This function provides a detailed analysis of expensive SQL statements. For more information, see The EXPLAIN Function [page 43]. To display the ABAP source program where the statement was defined, choose an entry in the list of ABAP reports. An editor screen appears, which contains the related source.
282
September 2011
10 Diagnostics
Caution Since all trace data remains permanently in the database, you should delete obsolete data before starting a new trace. To do so, choose the Delete pushbutton in the global toolbar. End of the caution. For information about activating the cumulative SQL trace function, see SAP Note 139286.
September 2011
283
11 BW Administration
11 BW Administration
Note All BWrelated actions in the DBA Cockpit are only available if the system has been recognized as a BW system. A system is recognized as a BW system if the BI_Content component is installed. End of the note. The following sections provide information about: The MDC Advisor Health Checks of the BW System Administration and Monitoring of the Near-Line Storage (NLS) Database
More Information
For more information about the MDC advisor, see the Database Administration Guide: SAP NetWeaver Business Warehouse 7.0 and Higher Administration Tasks: IBM DB2 for Linux, UNIX, and Windows at: http://service.sap.com/instguidesnw Database-Specific Guides <Your SAP NetWeaver Release> Operations
284
September 2011
11 BW Administration
September 2011
285
11 BW Administration
Recommendation If you are using a single-partition database, InfoProviders and PSA tables are already created with a distribution key. We recommend that you check the distribution keys of your InfoProviders and PSA tables before you add new partitions to your database. End of the recommendation. o BW-Specific Layout Check This check verifies if both fact tables of an InfoCube are distributed over the same database partitions. In addition, it checks whether the column SID_0CALMONTH or SID_0FISCPER is filled correctly with data if MDC was defined for an InfoCube with one of these columns. You cannot use the BWSpecific Layout Check to check SAP BW compression. 3. Choose the Execute pushbutton. The Schedule Job dialog box appears. 4. Choose if you want the check to start immediately in the background or at a later date, and confirm your selection. After the jobs have finished, the results are displayed in the Results of Consistency Check area as follows: Type of Check Result o Description The MDC Space Check returns no error if one of the following applies: MDC is not defined for an InfoCube or DataStore object. MDC is defined for an InfoCube or DataStore object, and the additional disk space consumption with MDC is less or equal 33%. o The MDC Space Check returns a warning if MDC is defined for an InfoCube or a DataStore object, and the additional disk space consumption with MDC is more than 33% but less than 50%. The MDC Space Check returns an error if MDC is defined for an InfoCube or DataStore object, and the additional disk space consumption with MDC is more than 50%.
286
September 2011
11 BW Administration
Type of Check
Result
Description
Note If the MDC Space Check returns a warning or an error, you should change the defined MDC settings using the Repair pushbutton. End of the note. o The Distribution Key Check returns no errors if the distribution key of an InfoCube, DataStore object, or PSA table is correct. The Distribution Key Check returns an error if one of the distribution keys of an InfoCube, DataStore object, or PSA table is missing or is not correct or is not defined. If the database is not partitioned, you can repair the distribution key using the Repair pushbutton. Otherwise, we recommend that you redistribute the table. o The BW-Specific Layout Check returns no errors if both fact tables of an InfoCube are distributed over the same database partitions or if the column SID_0CALMONTH or SID_0FISCPER is filled correctly with data. The BW-Specific Layout Check returns warnings if both fact tables of an InfoCube are not distributed over the same database partitions or if the column SID_0CALMONTH or SID_0FISCPER is not correctly filled with data. If the distribution check of the fact tables returns warnings, we recommend that you redistribute the tables using DB6CONV. If the check of the column SID_0CALMONTH or SID_0FISCPER returns warnings, open a customer message on SAP Service Marketplace. Note
September 2011
287
11 BW Administration
You can monitor the background job in the DBA Planning Calendar. End of the note.
More Information
The DBA Planning Calendar [page 234]
288
September 2011
11 BW Administration
Process
1. You access the BW Administration NLS Configuration screen by calling the DBA Cockpit and choosing BW Administration NLS Configuration in the DBA Cockpit. The following table shows all NLS databases that are connected to your SAP NetWeaver BW system providing the following information: Column Connection Name DB Name DB Server Description Name of the database connection that is used to access the database Name of the assigned NLS database Name of the host where the NLS database is installed
Schema Name Name of the NLS database schema User Name Name of the database user that is used to connect to the NLS database Indicates if the configuration information, which is stored as part of the NLS connection, matches the BW system information Configuration Status If the Configuration Status is not correct, the BW system cannot use a NLS connection. To update the configuration of the NLS connection, choose the Update Configuration pushbutton. Indicates if the remote NLS database is available Connection Status Note By default, this field is empty. To check the connection status, choose the Test Connection pushbutton. End of the note.
September 2011
289
11 BW Administration
2. You maintain your NLS configurations using the following pushbuttons from the application toolbar: o o o To create a new NLS configuration entry, you choose Add. To change an existing NLS configuration, you choose Change. To delete an existing NLS configuration, you choose Delete.
To perform these tasks, the standard maintenance screen for the configuration of database connections is used. In all cases, a dialog box appears with a detailed maintenance screen for the database connection related to the NLS database. To check the availability of the NLS connection, choose the Test Connection pushbutton. The result of the connection test appears in the action message window in the lower half of the screen. Note If you create the NLS connection using directly the database connection maintenance functions without the NLS Configuration screen, the configuration data for the NLS connection is not created. Make sure that you use the NLS Configuration screen. End of the note.
More Information
Configuring Database Connections [page 28]
290
September 2011
11 BW Administration
The following table displays all InfoProviders that meet the specified selection criteria: Column InfoProvider Name Name of the InfoProvider Total size of all database tables of the InfoProvider in KB Note InfoProvider Size (KB) This size information is not newly calculated but retrieved from values that were previously provided by the REORGCHK for All Tables job and might therefore not be accurate. If you do not run the REORGCHK for All Tables job, the displayed value is -1. For more information, see Scheduling a REORGCHK for All Tables [page 253]. End of the note. NLS Connection NLS Name Name of the NLS connection Name of the NLS object Total size of all database tables of the InfoProvider in the NLS database Note NLS Size This value is only available if the NLS database has been configured and properly set up for monitoring in the DBA Cockpit. For more information, see Configuration of Systems for Remote Monitoring [page 22] and Scheduling a REORGCHK for All Tables [page 253]. End of the note. Query Enabled Indicates whether the BW queries of the InfoProvider have been enabled to read data from the NLS database Description
September 2011
291
11 BW Administration
3. On the Single Table Analysis screen in the Space task area, enter the schema and the table name of the NLS table in the appropriate fields. End of the caution.
More Information
Analyzing and Maintaining a Single Table [page 154]
292
September 2011
11 BW Administration
Process
1. You access the NLS Synchronisation Check screen by calling the DBA Cockpit and choosing BW Administration NLS Synchronisation Check . 2. You choose the NLS connection that you want to check from the drop-down list. The consistency of the NLS database and the BW system is checked. The output is displayed in the following table: Column Checked Object Description Tree structure with the name of the InfoProvider and of inconsistent archiving requests
Required Action Action that is required to repair the inconsistent request Description Target Timestamp Additional information about the required action Time stamp of the operation as contained in the BW system
The consistency check returns one of the following results: o The NLS database is consistent No further action is required. o The NLS database is not consistent and it can be resynchronized If the NLS connection is not consistent but it can be resynchronized, the archiving requests with problems are displayed for each InfoProvider as well as the corrective action to be taken. To resynchronize the NLS database, choose the Resynchronize pushbutton. o The NLS database is not consistent and it must be restored In this case, you have to restore the NLS database to the point in time that is displayed in column Target Timestamp (see above table). For more information about restoring a DB2 database, see the Database Administration Guide: SAP on IBM DB2 for Linux, UNIX, and Windows.
September 2011
293
11 BW Administration
More Information
Database Administration Guide: SAP on IBM DB2 for Linux, UNIX, and Windows at: http://service.sap.com/instguidesnw Database-Specific Guides <Your SAP NetWeaver Release> Operations
294
September 2011
12 Appendix
12 Appendix
This appendix contains detail information about the various snapshot options.
Average time in milliseconds required to read data from disk into the buffer pool Average time in milliseconds required to write data from the buffer pool to disk
Number of read accesses to data in the buffer pool Number of read accesses to data on disk (I/O)
Physical Reads
This value includes the number of physical reads that were performed synchronously (by the database manager agents) and asynchronously (by the I/O servers).
September 2011
295
12 Appendix
Field
Physical Writes
This value includes the number of physical writes that were performed synchronously (by the database manager agents) and asynchronously (by the I/O cleaners). Number of read accesses to data on disk (by agents)
Synchronous Reads Synchronous Writes Temporary Logical Reads Temporary Physical Reads Index Logical Reads
Number of write accesses to data on disk (by agents) Number of logical read requests that required I/O to get data pages into the temporary tablespace Number of physical read requests that required I/O to get data pages into the temporary tablespace
Number of read accesses to index data in the buffer pool Number of read accesses to index data on disk (I/O)
Physical Reads
Since index data is read only by the database manager agents, this value includes the number of synchronously read index pages. Number of write accesses to index data on disk (I/O)
Physical Writes
This value includes the number of physical writes that were performed synchronously (by the database manager agents) and asynchronously (by the I/O cleaners). Number of read accesses to index data on disk (by agents)
Synchronous Reads Synchronous Writes Temporary Logical Reads Temporary Physical Reads Note
Number of write accesses to index data on disk (by agents) Number of logical read requests that required I/O to get index pages into the temporary tablespace Number of physical read requests that required I/O to get index pages into the temporary tablespace
Data is read or written in pages. A page can be 4 KB, 8 KB, 16 KB, or 32 KB in size. Unless otherwise specified, no distinction is made between synchronous and asynchronous accesses.
296
September 2011
12 Appendix
Inserts
Maximum allowed size in KB for the package cache Size The package cache contains access plans. The maximum allowed size is determined by database configuration parameter PCKCACHESZ. Indicates percentage at which the data is read from the package cache, rather than directly from the hard disk Quality This is calculated using the following formula: (package cache lookups - package cache inserts) / package cache lookups *100
September 2011
297
12 Appendix
Description Number of times an application looked for a section in the package cache Total number of times that an access plan was not available for use and had to be loaded into the package cache Number of times that the package cache overflowed the bounds of its allocated memory Largest size reached by the package cache
298
September 2011
12 Appendix
Index Asynchronous Physical Reads Asynchronous Physical Writes Asynchronous Read Requests Note Data is read or written in pages. A page can be 4 KB, 8 KB, 16 KB, or 32 KB in size. End of the note. Number of index pages that were read asynchronously from disk and written to the buffer pool by the I/O servers (prefetch) Number of index pages that were written asynchronously from buffer pool to disk (I/O cleaners) Number of asynchronous index read requests
Direct Writes Average I/O per Request Direct Reads Direct Writes
Average number of requests to read directly from disk Average number of requests to write directly to disk
September 2011
299
12 Appendix
Total number of statistics collection activities for creating Number of Statistics Collections statistics by the system without table or index scan during During Query Compilation query compilation Time Spent During Query Compilation Number of Synchronously Collected Statistics Time Spent on Synchronous Statistics Collection Activities Total time spent on creating statistics by system without table or index scan during query compilation in milliseconds Total number of synchronous statistics collection activities during query compilation Total time spent on synchronous statistics collection activities in milliseconds
300
September 2011
12 Appendix
Field Escalations
Description
Number of times that locks have been escalated from several row locks to a table lock Lock Escalations If the maximum allowed length of the lock list is reached, row locks are converted to table locks to save space in the lock list. This process is called lock escalations Number of times that locks have been escalated from several row locks to one exclusive table lock, or the number of times an exclusive lock on a row caused the table lock to become an exclusive lock Exclusive locks are important to track since they can impact the concurrency of your data because other applications cannot access data held by an exclusive lock.
Locks Locks Currently Total number of locks currently held by the applications Held Number of deadlocks that have occurred Deadlocks Detected Deadlock situations are recognized and resolved automatically by the database. The database configuration parameter lock escalations determines when a lock wait situation is resolved. Number of times that a request to lock an object timed out instead of being granted Lock Timeouts The database configuration parameter lock escalations determines when a lock wait situation is resolved.
September 2011
301
12 Appendix
Total Log Available to Database Used by Database Maximum Space Used Amount of primary log space in bytes in the database that is not being used by uncommitted transactions Total amount of primary log space in bytes currently used in the database Maximum amount of primary log space used in bytes Node with the least amount of available log space in bytes Node with Least Available Space Note This field is only displayed for global snapshots over all partitions. End of the note. Application with Oldest Transaction Secondary Log Logs Currently Allocated Maximum Space Total number of secondary log files that are currently being used for the database Maximum amount of secondary log space used in bytes Application ID (that corresponds to the agent_id value from the application snapshot) of the application that has the oldest transaction
302
September 2011
12 Appendix
Field Used Log Pages Read Written Log Buffer Consumption LSN Gap
Description
Number of log pages read from disk Number of log pages written to disk
Percentage of log space held by dirty pages in relation to log space specified by parameter SOFTMAX Percentage of log space held that will have to be redone for crash recovery pages in relation to log space specified by parameter SOFTMAX
Restart Range
Log Buffer Quality Log Buffer Hit Ratio Ratio of log data read from the buffer in relation to log data read from disk Number of times that agents have to wait for log data to write to disk while copying log records into the log buffer Log Buffer Overflows This value is incremented per agent per incident. For example, if two agents attempt to copy log data while the buffer is full, then this value is incremented by two.
Log Buffer I/O Average Write Time/Page Average Write Time/IO Average Read Time/Page Average Read Time/IO Average time per page in microseconds required to write log data to disk Average time per I/O in microseconds required to write log data to disk Average time per page in microseconds required to read log data from disk Average time per I/O in microseconds required to read log data from disk
September 2011
303
12 Appendix
304
September 2011
12 Appendix
Field
Description Number of Data Definition Language (DDL) statements that were executed, for example, CREATE TABLE, CREATE VIEW, ALTER TABLE, and DROP INDEX Sum of the host execution times in seconds for all the statements that were executed Remaining part of the above elapsed time in microseconds
DDL
Number of COMMIT statements that have been attempted Number of ROLLBACK statements that have been attempted
Rollbacks
Automatic rollbacks caused by error situations or deadlocks are not included. Number of dynamic SQL statements attempted Number of static SQL statements attempted Number of attempted SQL statements that failed
Dynamic SQL Static SQL Failed SQL Hash Joins Total Total Hash Loops
Total number of hash joins executed Total number of times that a single partition of a hash join was larger that the available sort heap space Number of times that hash join data exceeded the available sort heap space Number of times that hash join data exceeded the available sort heap space by less than 10% Total number of hash joins that were throttled back by the sort memory throttling algorithm A throttled hash join is a hash join that was granted less memory than requested by the sort memory manager. A hash join is throttled back when the memory allocation from the shared sort heap is close to the limit set by database configuration parameter sheapthres_shr. This throttling significantly reduces the number of overflows over thesheapthres_shr limit in a system that is not properly configured. The data reported in this element only reflects hash joins using memory allocated from the shared sort heap.
Overflows
Small Overflows
Post Threshold
September 2011
305
12 Appendix
Total time in milliseconds required for all sort processes Average sort time in milliseconds
Total Sorts Total number of sorts that have been executed Sort Overflows If the storage area allocated for sorting is not large enough, a sort overflow occurs. The hard disk is then used temporarily.
Active Sorts Number of sorts in the database that currently have a sort heap allocated Total number of sorts that were throttled back by the sort memory throttling algorithm. A throttled sort is a sort that was granted less memory than requested by the sort memory manager. A sort is throttled back when the memory allocation for sorts is close to the limit set by database configuration parameter sheapthres_shr. This throttling significantly reduces the number of overflows over sheapthres_shr limit in a system that is not properly configured. The data reported in this element only reflects sorts using memory allocated from the shared sort heap.
306
September 2011
12 Appendix
Number of times a buffer pool data page for an XML storage object (XDA) was physically written to disk by either an asynchronous page cleaner, or a Physical Writes prefetcher A prefetcher may have written dirty pages to disk to create space for the pages being prefetched. Read Requests Number of asynchronous read requests for XML storage object (XDA) data
September 2011
307
12 Appendix
Average time in milliseconds required to read data from disk into the buffer pool Average time in milliseconds required to write data from the buffer pool to disk
Physical Writes
308
September 2011
12 Appendix
Data Logical Reads Number of read accesses to data in the buffer pool Number of read accesses to data on disk (I/O) Physical Reads This value includes the number of physical reads that were performed synchronously (by the database manager agents) and asynchronously (by the I/O servers). Number of write accesses to data on disk (I/O) Physical Writes This value includes the number of physical writes that were performed synchronously (by the database manager agents) and asynchronously (by the I/O cleaners). Number of read accesses to data on disk (by agents)
Synchronous Reads Synchronous Writes Temporary Logical Reads Temporary Physical Reads Index Logical Reads
Number of write accesses to data on disk (by agents) Number of logical read requests that required I/O to get data pages into the temporary tablespace Number of physical read requests that required I/O to get data pages into the temporary tablespace
Number of read accesses to index data in the buffer pool Number of read accesses to index data on disk (I/O)
Physical Reads
Since index data is read only by the database manager agents, this value includes the number of synchronously read index pages. Number of write accesses to index data on disk (I/O)
Physical Writes
This value includes the number of physical writes that were performed synchronously (by the database manager agents) and asynchronously (by the I/O cleaners). Number of read accesses to index data on disk (by agents)
Number of write accesses to index data on disk (by agents) Number of logical read requests that required I/O to get index pages into the temporary tablespace
September 2011
309
12 Appendix
Number of physical read requests that required I/O to get index pages into the temporary tablespace
310
September 2011
12 Appendix
Direct Writes Average I/O per Request Direct Reads Direct Writes
Average number of requests to read directly from disk Average number of requests to write directly to disk
September 2011
311
12 Appendix
Number of times a buffer pool data page for an XML storage object (XDA) was physically written to disk by either an asynchronous page cleaner, or a Physical Writes prefetcher A prefetcher may have written dirty pages to disk to create space for the pages being prefetched. Read Requests Number of asynchronous read requests for XML storage object (XDA) data
312
September 2011
12 Appendix
Average time in milliseconds required to read data from disk into the buffer pool Average time in milliseconds required to write data from the buffer pool to disk
Number of read accesses to data in the buffer pool Number of read accesses to data on disk (I/O)
Physical Reads The value includes the number of physical reads that were performed synchronously (by the database manager agents) and asynchronously (by the I/O servers).
September 2011
313
12 Appendix
Number of write accesses to data on disk (I/O) Physical Writes The value includes the number of physical writes that were performed synchronously (by the database manager agents) and asynchronously (by the I/O cleaners). Number of read accesses to data on disk (by agents)
Temporary Logical Number of logical read requests that required I/O to get data pages into Reads the temporary tablespace Temporary Physical Reads Index Logical Reads Number of read accesses to index data in the buffer pool Number of read accesses to index data on disk (I/O) Physical Reads Since index data is read only by the database manager agents, this value includes the number of synchronously read index pages. Number of write accesses to index data on disk (I/O) Physical Writes The value includes the number of physical writes that were performed synchronously (by the database manager agents) and asynchronously (by the I/O cleaners). Number of read accesses to index data on disk (by agents) Number of physical read requests that required I/O to get data pages into the temporary tablespace
Temporary Logical Number of logical read requests that required I/O to get index pages into Reads the temporary tablespace Temporary Physical Reads Number of physical read requests that required I/O to get index pages into the temporary tablespace
314
September 2011
12 Appendix
September 2011
315
12 Appendix
Direct Writes Average I/O per Request Direct Reads Direct Writes
Average number of requests to read directly from disk Average number of requests to write directly to disk
316
September 2011
12 Appendix
Description Number of pages for XML storage objects (XDA) that have been requested from the buffer pool (logical reads) for temporary tablespaces. Number of pages for XML storage objects that have been (XDA) read from the tablespace containers (physical reads) for temporary tablespaces
Number of XML storage object (XDA) data pages that have been read in from the tablespace containers (physical reads) by asynchronous engine dispatchable units (EDUs) for all types of tablespaces.
Number of times a buffer pool data page for an XML storage object (XDA) was physically written to disk by either an asynchronous page cleaner, or a Physical Writes prefetcher A prefetcher may have written dirty pages to disk to create space for the pages being prefetched. Read Requests Number of asynchronous read requests for XML storage object (XDA) data
September 2011
317
12 Appendix
Current status of the application. Possible values are: Database Connect Pending The application has initiated a database connection but the request has not yet completed. Database Connect Completed The application has initiated a database connection and the request has completed. Unit of Work Executing The database manager is executing requests on behalf of the unit of work. Unit of Work Waiting The database manager is waiting on behalf of the unit of work in the application. This status typically means that the system is executing in the application's code. Lock Wait The unit of work is waiting for a lock. After the lock is granted, the status is restored to its previous value. Status Commit Active The unit of work is committing its database changes. Rollback Active The unit of work is rolling back its database changes. Recompiling The database manager is compiling an SQL statement or precompiling a plan on behalf of the application. Request Interrupted An interrupt of a request is in progress. Database Disconnect Pending The application has initiated a database disconnect but the command has not yet completed executing. The application might not have explicitly executed the database disconnect command. The database manager disconnects from a database if the application ends without disconnecting. Transaction prepared The unit of work is part of a global transaction that has entered the
318
September 2011
12 Appendix
prepared phase of the two-phase commit protocol. Transaction Heuristically The unit of work is part of a global transaction that has been heuristically committed. Transaction Heuristically Rolled Back The unit of work is part of a global transaction that has been heuristically rolled-back. Transaction Ended The unit of work is part of a global transaction that has ended but has not yet entered the prepared phase of the two-phase commit protocol. Creating Database The agent has initiated a request to create a database and that request has not yet completed. Restarting Database The application is restarting a database to perform crash recovery. Restoring Database The application is restoring a backup image to the database. Backing Up Database The application is performing a "fast load" of data into the database. Data Fast Load The application is performing a "fast load" of data into the database. Data Fast Unload The application is performing a "fast unload" of data from the database. Wait to Disable Tablespace The application has detected an I/O error and is attempting to disable a particular tablespace. The application has to wait for all other active transactions on the tablespace to complete before it can disable the tablespace. Quiescing a Tablespace The application is performing a QUIESCE TABLESPACE request. Wait for Remote Partition The application is waiting for a response from a remote partition in a partitioned database instance.
September 2011
319
12 Appendix
Agent PID Client Information Client user ID that is generated by the transaction manager and provided to the server, if the sqleseti API is used For ABAP systems: name of the SAP user Workstation Identifies the clients system or workstation (for example, CICS EITERMID) if the sqleseti API was used in this connection. Identifies the server transaction program performing the transaction if the sqleseti API was used in this connection. For ABAP systems: name of the SAP transaction The data passed to the target database for logging and diagnostic purposes, if the sqleseti API was used in this connection For ABAP systems: name of the program executing the SQL statement Process ID of an SAP work process that made the connection to the database
User ID
Application
Accounting
320
September 2011
12 Appendix
Number of agents participating in this application (high-water mark) Number of agents removed from this application and subsequently used by another application This only happens if the agent was not busy.
Times Used by Agent(s) User CPU Time System CPU Time Idle Time Total user CPU time in seconds consumed by agents Total system CPU time in seconds consumed by agents Total idle time in seconds
September 2011
321
12 Appendix
Displays information about memory pools allocated to this application The table contains the following columns: Partition Partition number PID Process ID (UNIX) or thread ID (Windows) of the agent Memory Pools Allocated to Agent Pool ID Type of memory pool Current Size (KB) Current size of the memory pool High-Water Mark (KB) Largest size of the memory pool since its creation Configured Size (KB) Configured size of the memory pool
322
September 2011
12 Appendix
Currently Associated Number of agents currently associated with the application Associated with This Appl. Number of agents participating in this application (high-water mark) Number of agents removed from this application and subsequently used by another application This only happens if the agent was not busy. Times Spent by Agent(s) User CPU Time System CPU Time Idle Time Waited for Prefetch Total user CPU time in seconds consumed by agents Total system CPU time in seconds consumed by agents Total idle time in seconds Total time in milliseconds waited for prefetch
September 2011
323
12 Appendix
This table displays information about all agents that are the coordinator agent itself or agents working as a subagent for the coordinator agent: Partition Partition number Agent Type Type of the agent Nesting Level Nesting level of the agent Entity Entity of the agent State Assigned Agents Indicates whether an agent is associated or active TID Thread ID of the agent Service Class Service class the agent is assigned to Event Type Type of event that was last processed by the agent Event Object Object of the event that was last processed by the agent Event State State of the event that was last processed by the agent
324
September 2011
12 Appendix
September 2011
325
12 Appendix
Average Time Physical Reads Average time in milliseconds required to read data from disk into the buffer pool Average time in milliseconds required to write data from the buffer pool to disk
Number of read accesses to data in the buffer pool Number of read accesses to data on disk (I/O)
Physical Reads
The value includes the number of physical reads that were performed synchronously (by the database manager agents) and asynchronously (by the I/O servers). Number of write accesses to data on disk (I/O)
Physical Writes
It includes the number of physical writes that were performed synchronously (by the database manager agents) and asynchronously (by the I/O cleaners). Number of logical read requests that required I/O to get data pages into the temporary tablespace Number of physical read requests that required I/O to get data pages into the temporary tablespace
Number of read accesses to index data in the buffer pool Number of read accesses to index data on disk (I/O)
Physical Reads
Since index data is read only by the database manager agents, this value contains the number of synchronously read index pages. Number of write accesses to index data on disk (I/O)
Physical Writes
The value includes the number of physical writes that were performed synchronously (by the database manager agents) and asynchronously (by the I/O cleaners). Number of logical read requests that required I/O to get index pages into the temporary tablespace Number of physical read requests that required I/O to get index pages into the temporary tablespace
326
September 2011
12 Appendix
Direct Writes Average I/O per Request Direct Reads Direct Writes
Average number of requests to read directly from disk Average number of requests to write directly to disk
September 2011
327
12 Appendix
Temporary Data Logical Reads Physical Reads Asynchronous I/O Physical Reads Number of XML storage object (XDA) data pages that have been read in from the tablespace containers (physical reads) by asynchronous engine dispatchable units (EDUs) for all types of tablespaces. Number of pages for XML storage objects (XDA) that have been requested from the buffer pool (logical reads) for temporary tablespaces. Number of pages for XML storage objects that have been (XDA) read from the tablespace containers (physical reads) for temporary tablespaces
Number of times a buffer pool data page for an XML storage object (XDA) was physically written to disk by either an asynchronous page cleaner, or a Physical Writes prefetcher A prefetcher may have written dirty pages to disk to create space for the pages being prefetched. Read Requests Number of asynchronous read requests for XML storage object (XDA) data
328
September 2011
12 Appendix
Number of times that locks have been escalated from several row locks to a table lock Lock Escalations If the maximum allowed length of the lock list is reached, row locks are converted to table locks to save space in the lock list. This process is called "lock escalation".
Number of times that locks have been escalated from several row locks to one exclusive table lock, or the number of times an exclusive lock on a row Exclusive Lock caused the table lock to become an exclusive lock Escalations Exclusive locks are important to track since they can impact the concurrency of your data because other applications cannot access data held by an exclusive lock. Locks Locks Currently Total number of locks currently held by the application Held Number of deadlocks that have occurred Deadlock situations are recognized and resolved automatically by the database The database configuration parameter DLCHKTIME determines when a lock wait situation is resolved. Number of times that a request to lock an object timed out instead of being granted Lock Timeouts The database configuration parameter LOCKTIMEOUT determines when a lock wait situation is resolved. Lock Timeout Value of the database configuration parameterLOCKTIMEOUT
Deadlocks Detected
September 2011
329
12 Appendix
Description
When a detailed deadlock event monitor with history is running, this element reports the number of bytes being used from the database monitor heap (HON_HEAP_S) to keep track of the statement history list entries.
DDL
Statements Attempted COMMITs Number of COMMIT statements that have been attempted Number of ROLLBACK statements that have been attempted Rollbacks Automatic rollbacks caused by error situations or deadlocks are not included. Number of dynamic SQL statements attempted
Dynamic SQL
330
September 2011
12 Appendix
Field Static SQL Failed SQL Hash Joins Total Total Hash Loops
Description Number of static SQL statements attempted Number of attempted SQL statements that failed
Total number of hash joins executed Total number of times that a single partition of a hash join was larger than the available sort heap space Number of times that hash join data exceeded the available sort heap space Number of times that hash join data exceeded the available sort heap space by less than 10%
Overflows
Small Overflows
September 2011
331
12 Appendix
Inserts
Overflows
Indicates percentage at which the data is read from the package cache, rather than directly from the hard disk Quality This is calculated using the following formula: (package cache lookups - package cache inserts) / package cache lookups * 100 Lookups Inserts Number of times an application looked for a section in the package cache Total number of times that a request section was not available for use and had to be loaded into the package cache
332
September 2011
12 Appendix
Log Space Used Log space used in bytes in most recent unit of work Completion Status Previous Unit of Work Stop Time Previous time that the most recent unit of work completed, which occurs when the database changes are committed or rolled back Completion status of last transaction
September 2011
333
12 Appendix
Field Rows Rows Read Rows Written Sort Sort Overflow Total Sort Time Number of Statement Sorts Data Logical Reads Physical Reads
Description
Number of rows read from the table Number of rows changed (inserted, deleted, or updated) in the table
If the storage area allocated for sorting is not large enough, a sort overflow occurs. The hard disk is then temporarily used. Total time for all sort processes Total number of sorts that have been executed
Number of read accesses to data in the buffer pool Number of read accesses to data on disk
Temporary Logical Number of logical read requests that required I/O to get data pages into Reads the temporary tablespace Temporary Physical Reads Index Logical Reads Physical Reads Number of read accesses to data in the buffer pool Number of read accesses to data on disk Number of physical read requests that required I/O to get data pages into the temporary tablespace
Temporary Logical Number of logical read requests that required I/O to get index pages into Reads the temporary tablespace Temporary Physical Reads Number of physical read requests that required I/O to get index pages into the temporary tablespace
334
September 2011
12 Appendix
September 2011
335
12 Appendix
Field COMPILE
Description
Blocking Cursor
Indicates if the statement that is executed is using a blocking cursor (YES) or not (NO) If data is transferred in blocks and not row by row, the performance of the corresponding query is improved. Text of dynamic SQL statement that was being processed when the snapshot was taken
Statement It can also be the text of the statement that was most recently processed if no statement was being processed at the time when the snapshot was taken. If a statement is displayed, you can choose EXPLAIN to list the access plan for the statement execution. This function provides a detailed analysis of expensive SQL statements. To show the current access plan of the statement, you can choose the EXPLAIN pushbutton. To display the ABAP source code from where the SQL statement was executed, you can choose the Show Source pushbutton. Note The Show Source pushbutton is only available if the following applies: The SAP system on which the currently monitored database is running is an ABAP system. The SAP system has the minimum Support Package level as described in SAP Note 1493490. End of the note.
More Information
The EXPLAIN Function [page 43]
336
September 2011
12 Appendix
High-Water Mark Largest size reached by the private workspace Shared Workspace Lookups Inserts Overflows Number of times an application looked for an SQL section in the agent's shared workspace Inserts of SQL sections by an application into the shared workspace Number of times that the shared workspace overflowed the bounds of its allocated memory
September 2011
337