h13827 Unisphere Central v4 WP
h13827 Unisphere Central v4 WP
h13827 Unisphere Central v4 WP
0 SP3
Next-Generation Storage Monitoring
ABSTRACT
This white paper introduces EMC Unisphere Central, a network
application that enables administrators to remotely monitor multiple Unity
Family, UnityVSA, VNX Series, and VNXe, EMC CLARiiON CX4
Series and vVNX Series storage systems residing on a common network.
This paper provides a detailed description of how to use this leading-edge
application.
December, 2016
WHITE PAPER
The information in this publication is provided as is. Dell Inc. makes no representations or warranties of any kind with respect to the
information in this publication, and specifically disclaims implied warranties of merchantability or fitness for a particular purpose.
Use, copying, and distribution of any software described in this publication requires an applicable software license.
Copyright 2016 Dell Inc. or its subsidiaries. All Rights Reserved. Dell, EMC, and other trademarks are trademarks of Dell Inc. or its
subsidiaries. Other trademarks may be the property of their respective owners. Published in the USA [12/16] [White Paper] [H13827.3]
Dell EMC believes the information in this document is accurate as of its publication date. The information is subject to change without
notice.
2
TABLE OF CONTENTS
MONITORING VNXE, VNX, CX4, VVNX, UNITY, AND UNITYVSA SYSTEMS WITH
UNISPHERE CENTRAL ............................................................................................................5
GETTING STARTED .................................................................................................................6
SYSTEM REQUIREMENTS .............................................................................................................. 6
STORAGE ARRAY REQUIREMENTS .............................................................................................. 7
DOWNLOAD AND DEPLOY THE UNISPHERE CENTRAL OVF TEMPLATE ................................. 8
CONFIGURE UNISPHERE CENTRAL SERVER.............................................................................. 8
ADD VNXE, VNX, CX4, VVNX, UNITY, AND UNITYVSA SYSTEMS TO UNISPHERE
CENTRAL ..................................................................................................................................9
VERIFICATION OF VNX AND CX4 SYSTEMS .............................................................................. 10
VERIFICATION OF VNXE, VVNX, UNITY, AND UNITYVSA SYSTEMS ........................................ 10
ALTERNATIVE METHOD TO ADD VNXE, VVNX, UNITY, AND UNITYVSA SYSTEMS TO
UNISPHERE CENTRAL .................................................................................................................. 12
REMOVE STORAGE SYSTEMS FROM UNISPHERE CENTRAL ................................................. 13
ADVANCED FEATURES........................................................................................................ 27
PREDICTIVE POOL CAPACITY MONITORING ............................................................................. 27
SINGLE SIGN-ON (SSO) ................................................................................................................ 28
PROFILE-BASED SYSTEM CONFIGURATION ............................................................................. 29
UNIFIED SEARCH .......................................................................................................................... 31
4
EXECUTIVE SUMMARY
In the world of storage management, storage administrators need a way to access information quickly about their environment
especially when it relates to critical systems. This can be a difficult task when an environment consists of multiple systems because
navigating to each individual system would be time consuming and identifying the systems that need immediate attention is critical.
To address these concerns, EMC introduced Unisphere Central, a network application that remotely monitors the status, activity,
and resources of multiple VNX Series, VNXe Series, CLARiiON CX4, vVNX, Unity Family and UnityVSA storage systems all
from a central location. The Unisphere Central server obtains aggregated status, alerts, host details, performance and capacity metrics,
and storage usage information from the systems. This allows administrators to take a quick look at their entire storage environment and
rapidly access storage systems which are in need of attention or maintenance.
INTRODUCTION
This white paper describes the Unisphere Central interface and explains how storage administrators can easily monitor multiple VNXe,
VNX, CLARiiON CX4, vVNX, Unity and UnityVSA storage systems. Also, it describes the Unisphere Central features and provides
general usage guidelines. For step-by-step instructions, refer to the Unisphere Central Online Help.
AUDIENCE
This white paper is intended for EMC customers, partners, and employees who are considering using Unisphere Central to monitor
VNXe, VNX, CLARiiON CX4, vVNX, Unity (supported on both Flash and Hybrid) and UnityVSA storage systems. EMC assumes the
reader is at least an IT generalist with experience as a system or network administrator.
TERMINOLOGY
Open Virtualization Format (OVF) A platform-independent, efficient, extensible, and open packaging and distribution format for
virtual machines.
Unisphere An interface for managing individual storage systems like VNX, VNXe, vVNX, Unity and UnityVSA.
Unisphere Central A Unisphere virtual application that collects and aggregates status information from appropriately configured
VNX, VNXe, CLARiiON CX4, vVNX Unity, and UnityVSA systems on a network. The Unisphere Central server is deployed as an
OVF template integrated within a VMware virtual environment.
Virtual Application (vApp) A container, such as a resource pool, that can contain one or more virtual machines. vApps also
share some functionality with virtual machines in that they can be powered on and off, and be cloned.
MONITORING VNXE, VNX, CX4, VVNX, UNITY, AND UNITYVSA SYSTEMS WITH
UNISPHERE CENTRAL
Unisphere Central enables you to:
Monitor up to 1000 VNXe, VNX, CX4, vVNX, Unity and UnityVSA systems for basic alerts and status information from a single
interface.
View aggregated alerts, system state, metrics storage, disk capacity, storage usage, and performance data for managed systems.
Control access to the monitoring interface by setting up local Unisphere Central users or integrating existing Lightweight Directory
Access Protocol (LDAP) enabled users and groups.
Organize logical views of all system types based on user-defined tags; for example, by location, type, or department.
Use Single Sign-On (SSO) functionality with VNXe3200 systems running code 3.1 or later, VNXe1600 systems, vVNX, Unity and
UnityVSA systems.
Initiate profile-based system configurations for VNXe3000 series systems running code 3.1.5 or later for VNXe1600 systems, and
for VNXe3200, Unity and UnityVSA systems.
The Unisphere Central environment consists of a Unisphere Central server running on an ESX/ESXi server (standalone or through
vCenter), VNXe, VNX, CX4, vVNX, Unity and UnityVSA storage systems, and a remote system to access the Unisphere Central server
(Figure 1).
GETTING STARTED
SYSTEM REQUIREMENTS
The following resources are required for deploying and using Unisphere Central:
At least one 64-bit ESXi server that meets the following requirements:
6
Table 1. ESXi System Requirements
RECOMMENDED
COMPONENT BASE CONFIGURATION
CONFIGURATION
CPU 2 4
MEMORY 3 GB > 8 GB
NETWORK INTERFACES 1 1
STORAGE 20 GB > 40 GB
The system requirements change when metrics collection is enabled in Unisphere Central. In this instance, please refer to Table 2
Table 3 in the APPENDIX A: PERFORMANCE METRICS for more information.
Note: High Availability configurations are only available when using ESXi servers are managed by a vCenter server. A minimum of two
ESXi servers are required.
DNS server (optional in IPv4 or IPv6 single-stack environments. Required in dual-stack IPv4/IPv6 environments).
An HTML5 compatible web browser or compatible device like a tablet to view the Unisphere Central HTML5 GUI. Supported web
browsers include:
- VNX for Block OE 05.33 and VNX for File OE 7.1 and later
o Release 30 04.30.000.5.525
Note: VNX File-only systems and VNX Gateway systems are not supported.
The EMC Unisphere Central Installation Guide, available from EMC Online Support, provides detailed installation instructions for
Unisphere Central.
Note: Unisphere Central installs with a default username, admin, and password, Password123#. When you first launch Unisphere
Central, you are required to change the administrator password.
In the Settings dialog, you can configure the following components in the Management Settings section (Figure 2):
Server Name (Network Settings Tab) The name of the Unisphere Central server (optional).
NTP servers (Time Servers (NTP) Tab) A protocol used to synchronize the system clock with other nodes on the network
(Highly recommended).
DNS servers (DNS Servers Tab) The network service that converts domain names to their corresponding IP addresses
(Optional in IPv4 or IPv6 single-stack environments. Required in dual-stack IPv4/IPv6 environments).
Security Policy (Security Policy Tab) Select either manual verification (least secure) or automatic verification (most secure,
default, and EMC recommended). Security policy settings apply to VNXe, vVNX, Unity, and UnityVS systems only.
o For manual verification, only the IP address of the Unisphere Central server must be configured on the storage systems
and the Unisphere Central administrator must verify the system manually before it can be monitored.
o For automatic verification, the server hash and challenge phrase configured here are provided to the storage system and
the identities of the Unisphere Central server and storage system are verified automatically when it connects to the
Unisphere Central server. The Unisphere Central server and storage systems always communicate through SSL
encrypted connections and authenticate each other by using X.509 certificates.
8
Figure 2. Settings Dialog
ADD VNXE, VNX, CX4, vVNX, UNITY, AND UNITYVSA SYSTEMS TO UNISPHERE
CENTRAL
Storage systems can be monitored by Unisphere Central without logging into each individual system. From the Systems > Storage
Systems page, click the Add icon and select either Add VNXe, Add VNX, Add vVNX, Add CX4, Add Unity or Add UnityVSA
depending on the model of your storage system. This opens the Add Storage System window (Figure 3). In this window,
administrators can enter the IP address of one of their storage systems storage processors (SPs), verify the SP certificate for initial
communication, and provide admin credentials to add their storage system to Unisphere Central.
When adding a VNX or CX4 system, Unisphere Central discovers and adds all other systems included in the same local domain of the
system being added. To keep track of Unisphere domains, Unisphere Central assigns system-defined tags. For more information on
system-defined tags, please refer to the Tags section later in this document.
Administrators can also use the wizard to add Multiple Storage Systems of the same type to Unisphere Central consecutively. This is
done by selecting the Multiple Storage Systems option, and browsing to a text file (*.txt) or a comma-separated value file (*.csv)
containing a list of IP addresses (IPv4/IPv6). The list must be in the following format:
<SP_IP1>
<SP_IP2>
<SP_IP3>
<SP_IP4>
9
Figure 3. Add Storage System Window
A challenge phrase is a string that the Unisphere Central server uses to identify VNXe, vVNX, Unity and, UnityVSA systems during the
initial connection. This allows the Unisphere Central server to automatically confirm that this is a valid client request. When the initial
SSL connection is made from the storage system to the Unisphere Central server, the server sends its certificate chain to the storage
system. By providing this server hash string, the Unisphere Central server confirms its identity to the storage system.
In this way, the challenge phrase allows the Unisphere Central server to confirm the systems identity, and the server hash allows the
system to confirm the identity of the Unisphere Central server. This mutual identification is required only during the initial connection.
After the initial connection, the server and the storage systems use standard X.509 certificates for mutual authentication.
10
Figure 4. Configure Security Policy
If the security policy is set for Manual verification, administrators must manually verify the storage systems certificates just like for
VNX and CX4 systems. An unverified system displays a padlock icon in the Storage Systems list. Selecting one or multiple unverified
systems and clicking the Verify button brings up the Verify Storage Systems dialog (Figure 5).
11
ALTERNATIVE METHOD TO ADD VNXE, VVNX, UNITY, AND UNITYVSA SYSTEMS TO UNISPHERE
CENTRAL
Storage systems can also be added to Unisphere Central using Unisphere.
Log into the VNXe/vVNX/Unity/UnityVSA
o For VNXe/vVNX select Settings > Management Settings, and click the Network tab.
In the Unisphere Central Configuration section (Figure 6) for VNXe and (Figure 7) for Unity
o For VNXe systems select the Configure this VNXe (system) for Unisphere Central checkbox, and type the Unisphere
Central IP address.
o For Unity systems select the Configure this storage system for Unisphere Central checkbox, and type the Unisphere
Central IP address.
If you set the security policy on the Unisphere Central server to Automatic, select the Use additional security information from
my Unisphere Central checkbox.
If you set the security policy on the Unisphere Central server to Manual, type only the IP address of Unisphere Central.
12
Figure 7. Unisphere Central Configuration in Unisphere for Unity
DASHBOARD PAGE
The Dashboard page displays customizable view blocks for all monitored storage systems (Figure 9). Unisphere Central enables users
to create, configure, and manage multiple dashboards. Each dashboard appears as a tab in the dashboard window and displays a set
of views.
13
Figure 9. Dashboard Page
The Dashboard page can be configured to display multiple dashboards containing any combination of the following view
blocks:
Alerts Displays a summary of alerts for all VNXe, vVNX, VNX, CX4, Unity, and UnityVSA systems that are monitored by
Unisphere Central (Figure 10). Users can also customize the name of the view block, display the number of alerts within a specific
time range, and display alerts with a specific tag. To customize, users need to hover over the view block and click Configure icon
(gear icon). Clicking an alert icon brings the user to the Alerts page which is automatically filtered based on what is clicked.
Capacity Provides a graphical summary of the storage capacity provisioned/used on the storage systems that Unisphere Central
monitors (Figure 11). The storage capacity provisioned/used on the systems is represented as parts of a donut chart. Tooltips
provide the value of each part of the chart by hovering over them. Users can also customize the name of the view block, display
capacity of systems with a specific tag, as well as show capacity based on storage resource (i.e. File Systems, LUNs, etc).
14
Figure 11. Capacity View Block
Pools Provides a snapshot view of 5 or 10 pools Unisphere Central monitors that have the most or least available size or
percentage of available size (Figure 12). Users can change the name of the view block, customize how many pools are shown (5
or 10), show pools from systems with a specific tag, and change the chart type (most available, least available, most percentage,
or least percentage). Users can also click a specific pool which brings them to the Pools page with the specified pool automatically
highlighted.
Pools Running out of Space Provides a summary of the pools that are running out of space with an estimated time until each
pool is full within a week, month, and quarter (Figure 13). Users can customize the name of the view block and display pools from
systems with a specific tag. Note that users will only be able to see pool out of space information when Unisphere Central has
metrics collection storage available and metrics collection is enabled on the storage systems. Clicking one of the categories brings
the user to the Pools page which is filtered based on what is clicked.
Health & Inventory Shows the health state of all storage systems currently monitored by Unisphere Central including some of
the resources available on the systems (Figure 14). Users can customize the name of the view block as well as display systems
with a specific tag. Clicking a category brings the user to the specified page with applicable filters.
15
Figure 14. Health & Inventory View Block
Tier Capacity Displays a summary of the free and used pool capacity in each tier of the storage pools (Figure 15). Users can
customize the name of the view block and filter the systems by a specific tag.
The Storage Systems page also allows users to filter the list of managed systems by table column content by clicking the Filter icon .
For example, users can filter systems based on certain severity levels like Critical and Major Problem. This is useful when there are
hundreds of systems in an environment and an administrator wants to find the systems that need attention in a timely manner.
The Export icon allows users to export the list of systems to a CSV file. Lastly, the Customize icon allows users to add more columns
of different categories for personalized reporting.
16
Figure 16. Storage Systems Page
TAGS
Storage systems monitored by Unisphere Central can have tags applied to them to help organize and identify the systems in a quick
and efficient manner. This is done by highlighting one or multiple systems, clicking the Tags > Apply Tags button, creating/choosing
the wanted tags, and clicking Apply. After a user assigns a tag to a managed system, they can filter systems by tags in the storage
systems list or in the customizable view blocks on the Dashboard page.
To keep track of Unisphere domains for VNX and CX4 systems, Unisphere Central automatically assigns system-defined tags to every
VNX and CX4 system. System-defined tags cannot be renamed, removed from any system, or deleted from Unisphere Central. These
tags use the following format:
User-defined tags can be managed using the Tags > Manage Tags button. In the Manage Tags dialog, users can create new tags,
rename tags, and delete tags (Figure 17).
PROFILES PAGE
The Profiles page allows administrators to create configuration profiles of supported VNXe, Unity, and UnityVSA systems and push
those profiles onto other VNXe, Unity, and UnityVSA systems so that they have the same settings (Figure 18). This is useful for
17
administrators with many storage systems that are about to be deployed and need the same configuration. The page allows users to
create, view, delete, apply, export, and import configuration profiles. Profiles can also be created and applied to systems through the
Storage Systems page using the More Actions button.
For more information on the profile-based system configuration feature, see the ADVANCED FEATURES section in this document.
STORAGE MENU
The Storage menu provides detailed information on storage use and configuration of the storage environment for the systems being
monitored by Unisphere Central. The menu includes separate pages for each type of supported storage resource including Storage
Pools, LUNs (Figure 19), LUN Groups, File Systems, VMware Datastores, Microsoft Hyper-V storage, and Microsoft Exchange storage.
Each storage resource page has the ability to be filtered based on the available columns and can be exported to a CSV file.
The frequency at which Unisphere Central collects storage and configuration data depends on the system type. For VNX and CX4
systems, Unisphere Central collects storage and configuration data once per hour unless it receives an alert, in which case Unisphere
Central polls for alerts every 5 minutes. For VNXe, vVNX, Unity, and UnityVSA systems, data is collected once per hour regardless of
alerts or health state changes. Data is retained until a monitored system or storage resource is removed.
18
Detailed information on each instance of a storage resource is also available in the following categories: general information,
associated hosts, storage capacity, and storage pool utilization and associated disks.
HOSTS PAGE
The Hosts page provides a list of hosts configured on the storage systems monitored by Unisphere Central (Figure 20). The
information displayed for each host includes the health state, host name, system, description, network address, initiators, and operating
system. Host data is collected once per hour and retained until a monitored host or storage system is removed. The host list can be
filtered based on column content by clicking the Filter icon. Also, the list can be exported to a CSV file.
ALERTS PAGE
The Alerts page provides a list of aggregated alerts from all storage systems being monitored by Unisphere Central (Figure 21). Users
can use alerts to determine the source of an issue, symptoms and cause of the issue, and actions that can be done to resolve it. Any
actions taken to resolve an alert must be performed directly on the system on which the alert was reported. The information displayed
for each alert includes:
Severity level
Log message
Description
The alerts can be filtered by column content and the entire list can be exported to a CSV file by clicking the Export icon.
19
Figure 21. Alerts Page
LOGS PAGE
The Logs page provides a list of log messages/events that Unisphere Central generates to record errors, commands, and other
information (Figure 22). The information displayed for each log includes:
Severity level
Source
Description
The logs/events can be filtered by column content and the entire list can be exported to a CSV file.
20
Figure 22. Logs Page
JOBS PAGE
The Jobs page allows users to view information for Unisphere Central jobs including those that are active and those that are completed
or failed (Figure 23). To quickly determine the number of active jobs (those queued or running) and view jobs progress, use the Jobs
icon in the bottom status bar. The information displayed for each job includes:
State
Percentage complete
Action
Description
The jobs can be filtered by column content and the entire list can be exported to a CSV file.
21
Figure 23. Jobs Page
PERFORMANCE PAGE
The Performance page allows users to view and interact with charts that display historical performance data for the storage systems
currently being monitored by Unisphere Central. Figure 24 illustrates an example of a performance metrics chart. Users can use the
charts to analyze storage system performance and diagnose any performance issues, such as:
Compare changes in performance across multiple metrics like network traffic, bandwidth, and throughput
Use preset or customized time ranges to view data collected within a specified time period
Before performance metrics can be seen, metrics collection must be enabled by allocating space for storing the collected metrics data.
This is done using the Create Metrics Storage window available in the Settings dialog on the Metrics tab. Also, users can expand
metrics storage by using the Expand Metrics Storage dialog. To display and analyze metrics data from VNX and CX4 systems, the
system must have statistics logging enabled. The VNX Unisphere Online Help describes how to enable statistics logging. For additional
metrics information and metrics storage sizing recommendations, refer to APPENDIX A: PERFORMANCE METRICS.
22
Figure 24. Performance Metrics Page (Average CPU Utilization Chart)
Metrics are shown in Unisphere Central using line charts which display the aggregate information for a single metric statistic. The line
chart shows the metrics value along the y-axis and the user-specified time range on the x-axis. Users can hover over a data point on
the line chart which displays a tool tip with the time, date, and measurement associated with that data point. Hovering over a data point
on one chart enables tooltips for all other displayed charts (Figure 24).
The default time range displayed is All, and users can click other time range values (1 day, 1 week, 1 month, 6 months) using the links
at the top of the page. A custom link is available to enter in specific start and end dates for viewing. Also, a navigator bar can be seen
near the top of the page which allows ease of time control and affects all displayed charts. Lastly, next to the navigator bar is a refresh
button which gathers the latest captured data to display in the charts.
Users can also break down the aggregated data into more detailed views by selecting applicable data contributors on the Configure
dialog. The dialog can be found by clicking the gear icon when hovering over a specific chart. Figure 25 illustrates an Average CPU
Utilization chart for which two contributors (SPA and SPB) have been selected. Each contributor shown is displayed as a different color
line and users can quickly remove and add each contributor by clicking on its name in the legend. This filter process allows users to
zero in on a subset of data. When a filter is applied to a line chart, the system redraws the chart to only show the metric values that
reflect the filter criteria.
23
Figure 25. Average CPU Utilization Chart with Selected Contributors
SETTINGS MENU
The Settings menu, seen by clicking the Settings icon, allows users to configure Unisphere Central settings. The menu contains the
following tabs under the Management Settings and Users & Groups tabs:
Network Settings Set up the server name and network settings for the Unisphere Central server.
Time Servers (NTP) Set up Network Time Protocol (NTP) settings to synchronize time with application hosts. The accuracy of
time on the Unisphere Central server is important to proper function of Unisphere Central. It is highly recommended to configure an
NTP server during the install process.
DNS Servers Set up Domain Name Servers (DNS) to resolve host names on a network. DNS Server(s) are required for dual-
stack IPv4/IPv6 environments.
Security Policy Configure the security policy to specify how storage systems are verified when connecting to Unisphere Central.
Automatic verification automatically verifies storage systems when they connect to Unisphere Central using a preconfigured
security certificate hash and passphrase. Manual verification requires users to manually verify connected storage systems.
Metrics Configure and manage metrics storage (Figure 26). Use this tab to do the following:
o Start metrics collection if collection was not automatically started after storage creation
24
Figure 26. Metrics Configuration Tab
Users & Groups Create, modify, and delete users and user groups. The user roles available for Unisphere Central are operator
and administrator. Note that a user must have administrator privileges to see and manage the users and groups list.
Directory Services Configure and manage LDAP server certificates through this page. LDAP helps centralize the management
of network authentication and authorization operations. Integrating Unisphere Central users into an existing LDAP environment
provides a way to control management access based on established user and group accounts within the LDAP directory. When
this is setup, a user can utilize an advanced feature called Single Sign-On (SSO). Please see the ADVANCED FEATURES section
for more information.
25
For upgrading system software, there can only be one update candidate file on the system at a time. EMC highly recommends that
users do not use the Unisphere Central server user interface during an update. Also, Unisphere Central may be temporarily
disconnected during the update, but automatically reconnects after the update. Before upgrading, users can use vSphere to create a
current snapshot of the Unisphere Central VM and use the snapshot to revert the Unisphere Central server to a previous version if
needed.
PREFERENCES DIALOG
The Preferences dialog, seen by clicking the Preferences icon, allows users to modify user preference settings including:
Preferences Set graphical interface user language, optimize the server for remote management access, and clear user cache
(Figure 28). Note that language packs need to be downloaded and installed before they can be chosen in this dialog.
Change Password Change the password of the currently logged in user account.
Logout Log out of the current user session. If Single-Sign On (SSO) is enabled, all other user sessions currently managed
through SSO are logged out as well.
SUPPORT DIALOG
The Support dialog, seen by clicking the Help icon, provides user access to the many support options available to Unisphere Central
including:
Support Provides instant access to online support information and communities. It is a central location for self-help resources
and it provides links to resources where users can learn about and get assistance with Unisphere Central.
Service EMC Unisphere Central Diagnose, troubleshoot, and repair the Unisphere Central server through this dialog (Figure
29). To access this dialog, users must type the service password which is configured during installation. The page provides tools to
collect service information to assist a service provider with a service request, export metrics data, change the service password,
enable/disable Secure Shell (SSH), and reboot the server.
26
Figure 29. Service EMC Unisphere Central Dialog
Online Help Central repository for all help topics and procedures. From this repository, users can locate a wide variety of
information including instructions for monitoring systems, detailed explanations, and instructions for servicing Unisphere Central.
Help for <Page> When navigating through the Unisphere Central interface, users can access this help link to be redirected
directly to a page-specific help topic in the Unisphere Central Online Help. Also, window dialogs provides a help icon that serves
the same purpose as the help link.
About Provides the current version number of the Unisphere Central server.
ADVANCED FEATURES
Unisphere Central is a useful tool for monitoring VNXe, vVNX, Unity and UnityVSA storage systems in a customers environment.
Beyond this basic functionality of the server, Unisphere Central offers administrators an extra feature set for different use cases. The
advanced features available to Unisphere Central are:
Unified search
27
Figure 30. Days to Full Column on the Pools Page
The feature helps users monitor the amount of space remaining on pools in their storage environment and plan for future capacity
needs. On the Dashboard page, users can also display a Pools Running out of Space view block which predicts pools being full in
the following ways:
To use the feature, users must have metrics collection enabled and metrics storage available through the Create Metrics Storage
wizard as well as metrics/statistics collection enabled on monitored storage systems. Also, Unisphere Central needs a minimum of 7
days of historical capacity metrics before making a prediction for a pool.
Unisphere Central uses a forecasting algorithm that identifies repeated patterns to predict future events. The algorithm starts
forecasting on the 8th day (after 7 days of historical capacity metrics data collection) with daily patterns and can identify future patterns
as data accumulates. A full quarter forecast would be available after 15 months of collected data. The forecasting algorithm runs once
per day.
Log into Unisphere Central, then select and launch Unisphere on a VNXe system, vVNX, Unity, or UnityVSA system without re-
authenticating.
Log into a storage system, open another browser window, and launch another Unisphere session for a different storage, by
entering the URL of that system, without having to re-authenticate.
Both Unisphere Central storage systems must be configured for the same AD/LDAP directory (Figure 31).
LDAP user must be directly mapped to a Unisphere role for both Unisphere Central and storage systems.
28
Each storage system must have SSO enabled.
Additionally, the feature provides a Single Sign-Off capability to users which allows the ability to log off of all systems in an SSO
environment from a single system. For more information on enabling SSO, please see the Unity Security Configuration Guide on EMC
Online Support.
DNS
NTP
Alerts
LDAP
Security
SMTP Server
Support
Location
User
29
Depending on the configuration, the following additional tabs are displayed:
Configuration profiles can be created, deleted, applied, imported, and exported from the Profiles page under the Systems menu.
Exported profiles are saved in a JSON format and can be imported to other Unisphere Central server instances. Profiles can also be
created and applied through the Storage Systems page through the More Actions menu.
The following are the requirements to use the profile-based system configuration feature:
To push FAST Cache settings, there must be available flash disks on the destination system.
o The destination system must have the available disks and necessary spare disks for the Storage Pool configurations.
Note: Storage Pool configuration capture is not supported for VNXe1600 systems.
30
Figure 32. Configuration Profile
UNIFIED SEARCH
Unisphere Central includes a unified search feature which allows users to search for objects based on string text criteria. This adds
convenience when trying to find a particular objects page or finding related objects for various purposes. The feature can be used by
clicking the Search icon at the top right of the Unisphere Central interface and inputting the search criteria (Figure 33).
31
To achieve HA, Unisphere Central takes advantage of VMware HA support provided by the VMware vSphere environment. VMware HA
is a feature of the host cluster. The vSphere environment provides several levels of high availability:
Host-level High Availability For the highest level of HA, Unisphere Central runs in a cluster of (at least) two hosts. This is the
recommended HA level for Unisphere Central services. At any given moment, the Unisphere Central Server VM runs on a single
host. When vSphere detects failure of the current Unisphere Central host (loss of heartbeat), or the failure of the Unisphere Central
VM, the Unisphere Central VM is restarted on the other host in the cluster. For this level of HA, the Unisphere Central VM must
reside on storage shared by both hosts.
VM-level High Availability When only a single host is available for Unisphere Central, HA still can be configured; however, it
does not protect against ESX failures. In this case, the Unisphere Central host must still be a part of the single-host cluster, which
is configured properly for HA. When vSphere detects a VM failure (loss of heartbeat), the VM is restarted on the same host.
Application-level High Availability Application-level HA in the Unisphere Central virtual machine is handled internally. No user
setup is required. However, if the Unisphere Central virtual machine fails more than three times in 10 minutes, it is not restarted
anymore. In that situation, the administrator needs to reboot the Unisphere Central virtual machine using the vSphere GUI. If that
does not help, the VM is apparently corrupted and needs to be restored from a previously backed-up virtual machine snapshot.
The EMC Unisphere Central Installation Guide, located on the EMC Online Support website, provides more information about HA
configurations in vSphere.
CONCLUSION
Unisphere Central was designed with the philosophy of keep it simple. It provides remote monitoring of the status, activity, and
resources of multiple VNXe, VNX, CX4, vVNX, Unity, and UnityVSA systems and the hosts connected to those resources making it
easier for administrators to oversee their environment.
Unisphere Centrals support ecosystem puts a world of resources at the administrators fingertips. Comprehensive online
documentation, help, training, and how-to videos are provided to expand the users knowledgebase and answer questions. All of these
features make Unisphere Central a powerful and easy-to-use tool.
32
REFERENCES
The following resources can be found on EMC Online Support:
Introduction to the VNX Series VNX5100, VNX5300, VNX5500, VNX5700, and VNX7500
33
APPENDIX A: PERFORMANCE METRICS
Unisphere Central gathers metrics from the monitored VNXe, VNX, CX4, vVNX, Unity and UnityVSA storage systems and displays
them in the Performance page.
The metrics functionality requires additional storage space to store the data collected from the storage systems. This storage space is
allocated from a VMware Datastore to the Unisphere Central vApp using a simple wizard.
Note: To display and analyze metrics data from VNX systems, the VNX system must have statistics logging enabled. The VNX
Unisphere Online Help describes how to enable statistics logging.
VCPU 2 4 8
MEMORY (GB) 4 6 8
STORAGE RESOURCES
10000 25000 50000
(LUNS, FILE SYSTEMS)
The total number of objects in small, medium, or large VNXe/Unity environments is substantially lower than the number of objects in the
corresponding VNX or CX4 environments. As a result, the recommended storage size provides more than enough metrics storage for
VNXe/Unity environments.
METRICS CONFIGURATION
Before users can use the metrics functionality, metrics collection must be enabled and users must allocate an initial amount of space for
storing the collected metrics using the Create Metrics Storage wizard located in the Metrics tab of the Settings dialog. During the
wizard, users can set the metrics data retention period. Unisphere Central allocates approximately 10 GB of storage immediately and
the remainder of the storage in a background process.
Note: You must allocate storage space before you can enable metrics collection.
Users can monitor metrics storage space utilization using the Metrics tab. If needed, users can expand the metrics storage by using
the Expand Metrics Storage dialog.
34
Table 5. Metrics Data Retention Period
PERFORMANCE METRICS
The Performance page allows users to view performance metrics for monitored VNXe, VNX, CX4, vVNX, Unity, and UnityVSA systems
in the environment. Table 6 and Table 7 list out the available VNXe 3.x/Unity and VNX performance metrics that are available,
respectively. For a complete list of metrics that can be collected by Unisphere Central, see Unisphere Central Online Help.
35
Table 6. Available VNXe /Unity Performance Metrics
RESOURCE Metric Description
CACHE* Cache Dirty Size Amount of data in the write cache, in MB, that has not yet been
flushed out to disks.
FAST Cache Dirty Ratio of dirty pages to total pages in the storage processor FAST
Pages Cache, expressed as a percent.
Total SP Cache Total number of I/O requests, in I/O per second, passing through
Throughput the storage processor cache.
CIFS Average CIFS I/O Average size of CIFS I/O requests, in KB, across all ports in the
Size storage system.
Total CIFS Total amount of CIFS I/O requests, in KB/s, across all ports in the
Bandwidth storage system.
Total CIFS Total number of CIFS I/O requests, in I/O per second, across all
Throughput ports in the storage system.
CPU Average CPU Total amount of processing cycles, as a percentage, across all
Utilization cores in the storage system SPs.
DISK* Average Disk Average number of disk I/O requests in the system queue, across
Queue Length all disks in the storage system.
Average Disk Average time spent completing disk I/O requests, in
Response Time microseconds, across all disks in the storage system, including
time spent in the queue.
Average Disk Average time spent completing disk I/O requests, in ms, across all
Service Time disks in the storage system, not including time spent in the queue.
Total Disk Total amount of disk I/O requests, in KB/s, across all disks in the
Bandwidth storage system.
Total Disk Total number of disk I/O requests, in I/Os per second, across all
Throughput disks in the storage system.
FC* Total Fibre Channel Total amount of Fibre Channel I/O requests, in KB/s, across all
Bandwidth ports in the storage system
Total Fibre Channel Total number of Fibre Channel I/O requests, in I/O per second,
Throughput across all ports in the storage system.
FILE SYSTEM Average File Average size of file system I/O requests, in KB, across all file
System I/O Size systems in the storage system.
Total File System Total amount of file system I/O requests, in KB/s, across all file
Bandwidth systems in the storage system.
Total File System Total number of file system I/O requests, in I/O per second,
Throughput across all file systems in the storage system.
ISCSI Total iSCSI Total amount of iSCSI I/O requests, in KB/s, across all ports in the
Bandwidth storage system.
Total iSCSI Total number of iSCSI I/O requests, in I/O per second, across all
Throughput ports in the storage system.
LUN Average LUN I/O Average size of LUN I/O requests, in KB, across all LUNs in the
Size storage system
Average LUN Average number of LUN I/O requests in the system queue across
Queue Length all LUNs in the storage system.
Average LUN Average time spent completing LUN I/O requests, in
Response Time microseconds, across all LUNs in the storage system.
Total LUN Total amount of LUN I/O requests, in KB/s, across all LUNs in the
Bandwidth storage system.
Total LUN Total number of LUN I/O requests, in I/Os per second, across all
Throughput LUNs in the storage system.
NFS Average NFS I/O Average size of NFS I/O requests, in KB, across all ports in the
Size storage system.
Total NFS Total amount of NFS I/O requests, in KB/s, across all ports in the
Bandwidth storage system.
Total NFS Total number of NFS I/O requests, in I/O per second, across all
Throughput ports in the storage system.
*Not available on UnityVSA
36
Table 7. Available VNX Performance Metrics
RESOURCE METRIC DESCRIPTION
CACHE Cache Dirty Ratio of cache, as a percentage, written but not flushed to disk
CPU CPU Utilization Total processing cycles, as a percentage, across all cores in
the storage processors utilized by the storage system
DISK Disk Bandwidth Total amount of disk I/O requests, in bytes per second, across
all disks in the storage system
Disk Errors Total number of disk errors encountered by all disks in the
storage system
Disk Response Average time, in ms, required for one request to pass through a
Time system component, including its waiting time
Disk Queue Average number of disk I/O requests in the system queue, in
Length MB/s, across all disks in the storage system
Disk Service Average time spent completing disk I/O requests, in ms, across
Time all disks in the storage system, not including time spent in the
queue
Disk Total number of disk I/O requests, in I/Os per second, across
Throughput all disks in the storage system
LUN/SP LUN Bandwidth Total amount of LUN I/O requests, in bytes per second, across
all LUNs in the storage system
LUN Response Average time, in ms, required for one request to pass through a
Time system component, including its waiting time
LUN Queue Average number of LUN I/O requests in the system queue
Length across all LUNs in the storage system
LUN Total number of LUN I/O requests, in I/Os per second, across
Throughput all LUNs in the storage system
PORT Port Bandwidth Total amount of I/O requests, in bytes per second, across all
ports in the storage system
Port Total number of I/O requests, in packets per second, across all
Throughput ports in the storage system
PROTOCOL/ Protocol Total number of protocol I/O requests, in I/Os per second,
OPERATION/ VERSION/ Throughput handled by the storage system
MOVER
VOLUME/MOVER File System Total amount of I/O requests, in bytes per second, across all
Bandwidth hosts associated with the storage system
File System Total amount of I/O requests, in packets per second, across all
Throughput hosts associated with the storage system
37