000-647 RPT

Download as pdf or txt
Download as pdf or txt
You are on page 1of 43

000-647

QUESTION: 1
What documents all artifacts that define the system, users, and tasks associated with the
target workload for the performance testing effort?

A. datapool document
B. product manual
C. test schedule guide
D. workload analysis document

Answer: D

QUESTION: 2
What is the recommended number of variables you should change at one time for tests that
will be compared?

A. all
B. four
C. none
D. one

Answer: D

QUESTION: 3
What are the most common measurement criteria collected during performance tests?

A. execution timestamp and passed verification points


B. object recognition and verification points
C. response times and throughput
D. tester names and test user groups

Answer: C

QUESTION: 4
Which type of performance test determines the conditions under which a system fails and
how it fails?

A. load
B. longevity and endurance
C. stress
D. volume

2 http://www.hotcerts.com
000-647

Answer: C

QUESTION: 5
What is the first step in workload definition?

A. clarify and document the goals of the performance-testing project


B. examine the production environment being tested
C. identify test variables and user groups
D. record a performance test to verify that the application is testable

Answer: A

QUESTION: 6
What is the goal of the workload analysis?

A. to clarify the objective of the workload interval


B. to ensure that tests performed represent real user activity
C. to gather statistics about the system under test
D. to identify the critical measurement points for each test

Answer: B

QUESTION: 7
You want to use test data in RPT custom code that comes from a server (for example, to
validate a response or to calculate a running total). What must you do to accomplish this?

A. create a datapool value that gets its value from a test reference for the server value, then
add this to the custom code argument list
B. create a reference for the server value at the beginning of the test, then add this to the
custom code argument list
C. create a reference for the server value anywhere before/above the point in the test where
the value is returned, then add this to the custom code argument list
D. create a reference for the server value at the footer of the test, then add this to the custom
code argument list

Answer: C

QUESTION: 8
You manipulate the execution of steps in a test by repeating sequences of a transaction. How
can you add further control?

3 http://www.hotcerts.com
000-647

A. add a Loop test element and then Events and Dependencies in the test suite
B. add do or for loop statements to the test source code and then use either the breakLoop()
or the continueLoop() method
C. add a Loop test element and then custom code classes with either the breakLoop() or the
continueLoop() method
D. add a Loop test element and then Loop Control test elements in the test editor

Answer: C

QUESTION: 9
What must you do if you want to record tests on a web browser other than Microsoft Internet
Explorer?

A. download and install the RPT IE plug-in for the supported web browser from
developerWorks
B. select Enable Environments for Testing from the Configure menu and add the web
browser as described in the product documentation
C. manually configure the browser to communicate to the server(s) via a SOCKS proxy as
described in the product documentation
D. manually configure the web browser to use the Microsoft JVM as described in the
product documentation

Answer: C

QUESTION: 10
How can you create a copy of an RPT test?

A. right-click the test; select Copy from the menu; enter a new name for the copied test
B. select the test; from the File menu, click ? Save As ; enter a new name for the copied test
C. switch to the Navigator view; select the source Java file (*.rec) of the test; from the File
menu, click ? Save As; enter a new name for the copied test
D. from the Navigator view, right-click the .recmodel file; select Generate Test; choose
Performance Test Generator; enter a new name for the copied test

Answer: D

QUESTION: 11
What is one way to modify or enhance a test programmatically?

A. add Java custom code classes to a test as well as directly to the test code (*.java)
B. add Java custom code classes to a test, but not directly to the test code (*.java)
C. directly modify the test code (*.java) by adding custom code classes to the appropriate

4 http://www.hotcerts.com
000-647

files
D. modify the default test superclass with custom code

Answer: B

QUESTION: 12
How can you make RPT tests more readable for others?

A. add comments during recording and enter Javadoc information for each test
B. label all blocks, add comments and use a naming convention with folders
C. label all timers, rename page titles and use the recorded think time
D. rename page titles, add comments, and use a naming convention with folders

Answer: D

QUESTION: 13
What must you do to change the connection security in an existing test from SSL to non-
secure (assuming the servers are reconfigured for this)?

A. use the Test Search feature to find and replace each connection string from http to https
B. create a new location for each server using an SSL connection and deselect the Enable
SSL checkbox in the server details window
C. go to a performance test schedule and deselect the Enable SSL checkbox in the Schedule
Element Details connection settings
D. go to the test editor and deselect the Enable SSL checkbox in the Connection Details
Authentication and Security pane

Answer: D

QUESTION: 14
What must you do if you want to reuse RPT tests on a server other than the one on which
they were recorded?

A. use the Test Search feature to find and replace the server host name(s) in each RPT test
B. select each page for each RPT test in the test editor and change the Target host to the
new server host name
C. create a new location and deployment for the new server host name(s) and add the RPT
tests as artifacts for the deployment
D. re-record the tests in the new environment because it is not possible to reuse a recorded
test on different servers

5 http://www.hotcerts.com
000-647

Answer: A

QUESTION: 15
You have an RPT test that does not indicate the correct number of pages in the actual web-
based application. An application goes through eight distinct pages but the RPT recorded test
only shows six pages. How can you correct this?

A. change the Performance Test Generation - Protocol preferences and re-generate the test
from the recorded session (.recmodel file)
B. change the HTTP Recorder Options and re-record the test
C. manually add the missing pages or remove the extra/redundant pages from the test editor
D. re-record the test and, while recording, click on New Page from the Recorder Control tab
where the pages were skipped

Answer: A

QUESTION: 16
Which technique of adding loops and conditionals to an RPT test provides the greatest
flexibility and is the least limited in its capability?

A. adding the loop and condition test elements in the test editor
B. adding the loop and condition test elements in a performance schedule
C. adding loops and conditional Java statements in custom code classes in the test
D. adding loops and conditional Java statements in the recorded session (*.rec)

Answer: C

QUESTION: 17
What is the fastest way to debug a test by validating data returned from a server?

A. run the test and then, in the Execution History, view the page in the Protocol Data
Browser tab and manually verify the displayed value
B. enable a content verification point (VP) for the value on the page where it is displayed
C. create a reference from the returned value and then validate the reference against a
datapool using a condition test element
D. create a reference from the returned value and then validate the reference against a
datapool using an if-then statement in custom code

Answer: B

QUESTION: 18
How can you pass values from one simulated user to another during a test?

6 http://www.hotcerts.com
000-647

A. create a global variable for each value in the test schedule and add references to both
store and retrieve the test values in the global variables
B. add custom code classes to the test using the IDataArea put() and get() methods to both
store and retrieve the test values
C. create a shared class for all tests and then use the IDataArea put() and get() methods to
both store and retrieve the test values
D. add custom code classes to the test using any Java file I/O method to write and read the
values to a persistent file

Answer: B

QUESTION: 19
What is one way to debug an RPT test by logging test values (server data, datapool
selections, etc.)?

A. add Log Entry test elements to the test in the test editor
B. use the Log Entry script support function while recording the test
C. add the test execution services reportMessage() method to the test source code
D. add custom code classes to the test using the reportMessage() method

Answer: D

QUESTION: 20
To validate correct test playback, you can enable Page Title Verification Points _____.

A. only if you have not renamed the page titles from their original recorded values
B. regardless of whether or not the page titles have been renamed
C. regardless of whether or not the page titles have been renamed, but only if each page title
has a unique name
D. regardless of whether or not the page titles have been renamed, but only if the page titles
match the expected page title values

Answer: B

QUESTION: 21
During test execution, the datapool framework is used to vary _____.

A. user names and passwords


B. data in a transaction flow
C. the contents in responses

7 http://www.hotcerts.com
000-647

D. request header variables

Answer: B

QUESTION: 22
Session IDs being used and maintained throughout a transaction is an example of _____.

A. a datapool with unique user names and passwords only


B. a datapool with unique user names, passwords, and sessions IDs
C. a combination of datapooling and correlation techniques
D. built-in auto correlation features

Answer: D

QUESTION: 23
Which is a challenge of test data management?

A. understanding the relationship of interrelated data needed for datapools and correlation
B. using the datapool import wizard when managing large amounts of data
C. developing representative and realistic tests, while minimizing the overhead introduced
by the test tool
D. managing the datapool data for all the iterations of the virtual testers across all agents

Answer: C

QUESTION: 24
Generally, applications that use client side logic require the relevant subsequent responses to
be correlated with a _____.

A. field reference
B. reference
C. variable from a datapool
D. custom code class

Answer: D

QUESTION: 25
What does each virtual user use during every loop when the open mode of a datapool is set to
private with no wrap (i.e. the wrap check box is unchecked)?

8 http://www.hotcerts.com
000-647

A. a unique row from the datapool because the test will not retrieve restart from the
beginning of the datapool when the end of the datapool is reached
B. a private view of the datapool with unique rows apportioned
C. a private view of a datapool with values that can be reused by other virtual users
D. a private view of a datapool assigned per agent machine

Answer: C

QUESTION: 26
By default, an element that has been substituted with a datapool variable appears in a test as
_____.

A. light blue text on a dark blue background


B. dark blue text on a light blue background
C. white text on a dark green background
D. dark pink text on a pink background

Answer: C

QUESTION: 27
The Test Data table for a page is used to _____.

A. view the locations of test elements


B. create substitutions for datapools
C. validate the URL encoded substitutions
D. view candidates for substitution

Answer: D

QUESTION: 28
What is the purpose of test data management for load testing?

A. to ensure representative and realistic tests where metrics can be gathered accurately
B. to ensure that data are not retrieved from the various cache lines in the system under test
C. to ensure that tests play back correctly after recording
D. to use different user names and passwords for each iteration

Answer: A

QUESTION: 29

9 http://www.hotcerts.com
000-647

Once a reference has been established, what is used to most efficiently find all subsequent
potential substitutions?

A. the Search and Replace tool


B. the Test Search tool searching all Requests
C. the "find more" and "go to" functions from the right-click submenu on a candidate
D. the Test Data preview pane under the test elements

Answer: B

QUESTION: 30
A customer recently purchased RPT. The playback tests appear to have passed in the Test
Execution History. However, upon closer inspection using the Protocol Data view, some
pages following the datapool substitution failed. What is the most likely cause of this
problem?

A. The datapool was opened in the wrong shared mode.


B. The datapool variable was substituted with the incorrect test element.
C. The datapool was not checked with a verification point.
D. The datapool value was substituted, but not fully correlated.

Answer: D

QUESTION: 31
What are the main performance schedule elements?

A. loops and delays


B. user groups and tests
C. selectors and comments
D. number of users and run duration

Answer: B

QUESTION: 32
What is the maximum recommended JVM heap size?

A. 1.9 GB
B. 2 GB
C. 80% of the physical memory
D. It is different for every supported OS.

10 http://www.hotcerts.com
000-647

Answer: A

QUESTION: 33
Which step is NOT required to validate and debug performance schedules?

A. remove any resource bottlenecks


B. validate initial response time measurements
C. run performance schedule in multi-user mode
D. find maximum user capacity of the system under test

Answer: D

QUESTION: 34
What is used to model a web page hit rate?

A. loops
B. delays
C. paced loops
D. random selectors

Answer: C

QUESTION: 35
The amount of memory consumed by the workbench JVM depends on the_____.

A. transaction rate
B. OS of the test agent
C. number of virtual users
D. amount of monitored data

Answer: D

QUESTION: 36
What do the sampling options in a schedule allow you to reduce in a large system test?

A. the CPU overhead


B. the total test duration
C. the number of virtual users
D. the required network bandwidth

11 http://www.hotcerts.com
000-647

Answer: A

QUESTION: 37
What simulates repeated user activity?

A. loops
B. repeated blocks
C. selectors
D. delays

Answer: A

QUESTION: 38
How do you control random user activity?

A. set an iteration rate


B. use additional custom code
C. set the number of iterations
D. define a weight for each test item

Answer: D

QUESTION: 39
What is the purpose of the runtime settings for Statistics?

A. to set the Execution History Log Level


B. to set the Problem Determination Log Level
C. to control the amount of data that is collected during a performance schedule run
D. to control the amount of detail that is available in the statistical data from a Performance
Test

Answer: C

QUESTION: 40
What limits the available memory for virtual users?

A. the JVM maximum heap size


B. the type of OS on the test agent
C. the CPU architecture (32-bit vs. 64-bit)
D. the amount of physical memory on the test agent

12 http://www.hotcerts.com
000-647

Answer: A

QUESTION: 41
Which is NOT included in the schedule elements details?

A. user group size


B. user load
C. execution history
D. schedule statistics

Answer: A

QUESTION: 42
What does weighting in a random selector allow you to control?

A. the frequency of user activities


B. the average think time within a selector
C. the number of iterations of a specific user activity
D. the total run time of a specific user activity

Answer: A

QUESTION: 43
The level of detail in the Execution History is dependent on _____.

A. log level
B. group size setting
C. the user think time limit
D. the number of users

Answer: A

QUESTION: 44
What is the correct way to achieve user pacing that emulates realistic user activity?

A. use the recorded think time


B. specify a fixed think time
C. vary the think time by a random percentage
D. increase or decrease the think time by a percentage

Answer: C

13 http://www.hotcerts.com
000-647

QUESTION: 45
Which is NOT a possible report customization?

A. rearrange items by specifying a sort order


B. insert items by adding new counters to the report
C. control the appearance of items by specifying the graph type
D. focus on items of interest by filtering data

Answer: A

QUESTION: 46
While running a performance test, the Server Health Detail report shows that pages are
failing as the load increases. You stop the run and examine the Test Execution History. It
also shows that pages are failing, but the page response Protocol Data view is blank. What
should you do?

A. examine the Test Agent error logs for the failed page error data
B. change the Execution History Log Level to "All" and re-run the test
C. leave everything the same and re-run the test, but this time do not stop it mid-way
D. change the Execution History Log Level to Secondary Test Actions and re-run the test

Answer: B

QUESTION: 47
How do you view Execution History during a run?

A. You edit the Agent location file to enable real-time transfer of Execution History data.
B. Execution History is transferred once the run completes. It cannot be viewed during a
run.
C. You select an Agent host under the appropriate schedule in the Performance Test Runs
view.
D. You set the Execution History Log Level to "Schedule Actions."

Answer: A

QUESTION: 48
It is common to exclude collected test data during ramp up and ramp down of a performance
test. What do you do to discard the last ten (10) minutes of a test for reporting purposes?

A. set the graphs X Max property to 600

14 http://www.hotcerts.com
000-647

B. set the graphs X Sliding Scale property to 600


C. set the graphs X Min property to the total run time minus 600
D. set the graphs X Max property to the total run time minus 600

Answer: D

QUESTION: 49
Performance Test results data can be exported from RPT in XML, HTML and CSV format.
Which view contains information that can be exported in CSV format?

A. Test Navigator
B. Performance Report
C. Performance Test Runs
D. Test Execution History

Answer: C

QUESTION: 50
You need to report the total server throughput for all traffic during a test run using the
default Test Report. Which counter is best to use?

A. Performance Report: Page Throughput Page Hit Rate


B. Page Element Report: Page Element Throughput Page Element Hit Rate
C. Transaction Report: Transaction Throughput Transaction Rate per Second
D. Page Element Report: Page Element Throughput Page Element Attempt Rate

Answer: B

QUESTION: 51
Which view is most useful for identifying the cause of failing page elements?

A. Test Execution History


B. Verification Points Report: Summary
C. Page Element Report: Server Health Detail
D. Page Element Report: Response vs. Time Detail

Answer: A

QUESTION: 52
What does the Performance Report: Response vs. Time Detail Graph show?

15 http://www.hotcerts.com
000-647

A. the average page response time per interval for each individual page vs. run duration
B. the maximum page response time per interval for each individual page vs. run duration
C. the measured page response time for each individual page vs. run duration
D. the average response time for all page elements during the run vs. run duration

Answer: A

QUESTION: 53
What documents all artifacts that define the system, users, and tasks associated with the
target workload for the performance testing effort?

A. datapool document
B. product manual
C. test schedule guide
D. workload analysis document

Answer: D

QUESTION: 54
Which type of performance test determines the conditions under which a system fails and
how it fails?

A. load
B. longevity and endurance
C. stress
D. volume

Answer: C

QUESTION: 55
What is the recommended number of variables you should change at one time for tests that
will be compared?

A. all
B. four
C. none
D. one

Answer: D

16 http://www.hotcerts.com
000-647

QUESTION: 56
What are the most common measurement criteria collected during performance tests?

A. execution timestamp and passed verification points


B. object recognition and verification points
C. response times and throughput
D. tester names and test user groups

Answer: C

QUESTION: 57
What is the first step in workload definition?

A. clarify and document the goals of the performance-testing project


B. examine the production environment being tested
C. identify test variables and user groups
D. record a performance test to verify that the application is testable

Answer: A

QUESTION: 58
What is the goal of the workload analysis?

A. to clarify the objective of the workload interval


B. to ensure that tests performed represent real user activity
C. to gather statistics about the system under test
D. to identify the critical measurement points for each test

Answer: B

QUESTION: 59
You manipulate the execution of steps in a test by repeating sequences of a transaction. How
can you add further control?

A. add a Loop test element and then Events and Dependencies in the test suite
B. add do or for loop statements to the test source code and then use either the breakLoop()
or the continueLoop() method
C. add a Loop test element and then custom code classes with either the breakLoop() or the
continueLoop() method
D. add a Loop test element and then Loop Control test elements in the test editor

17 http://www.hotcerts.com
000-647

Answer: C

QUESTION: 60
What must you do if you want to record tests on a web browser other than Microsoft Internet
Explorer?

A. download and install the RPT IE plug-in for the supported web browser from
developerWorks
B. select Enable Environments for Testing from the Configure menu and add the web
browser as described in the product documentation
C. manually configure the browser to communicate to the server(s) via a SOCKS proxy as
described in the product documentation
D. manually configure the web browser to use the Microsoft JVM as described in the
product documentation

Answer: C

QUESTION: 61
To validate correct test playback, you can enable Page Title Verification Points _____.

A. only if you have not renamed the page titles from their original recorded values
B. regardless of whether or not the page titles have been renamed
C. regardless of whether or not the page titles have been renamed, but only if each page title
has a unique name
D. regardless of whether or not the page titles have been renamed, but only if the page titles
match the expected page title values

Answer: B

QUESTION: 62
How can you make RPT tests more readable for others?

A. add comments during recording and enter Javadoc information for each test
B. label all blocks, add comments and use a naming convention with folders
C. label all timers, rename page titles and use the recorded think time
D. rename page titles, add comments, and use a naming convention with folders

Answer: D

QUESTION: 63
What must you do to change the connection security in an existing test from SSL to non-
secure (assuming the servers are reconfigured for this)?

18 http://www.hotcerts.com
000-647

A. use the Test Search feature to find and replace each connection string from http to https
B. create a new location for each server using an SSL connection and deselect the Enable
SSL checkbox in the server details window
C. go to a performance test schedule and deselect the Enable SSL checkbox in the Schedule
Element Details connection settings
D. go to the test editor and deselect the Enable SSL checkbox in the Connection Details
Authentication and Security pane

Answer: D

QUESTION: 64
You have an RPT test that does not indicate the correct number of pages in the actual web-
based application. An application goes through eight distinct pages but the RPT recorded test
only shows six pages. How can you correct this?

A. change the Performance Test Generation - Protocol preferences and re-generate the test
from the recorded session (.recmodel file)
B. change the HTTP Recorder Options and re-record the test
C. manually add the missing pages or remove the extra/redundant pages from the test editor
D. re-record the test and, while recording, click on New Page from the Recorder Control tab
where the pages were skipped

Answer: A

QUESTION: 65
What is one way to debug an RPT test by logging test values (server data, datapool
selections, etc.)?

A. add Log Entry test elements to the test in the test editor
B. use the Log Entry script support function while recording the test
C. add the test execution services reportMessage() method to the test source code
D. add custom code classes to the test using the reportMessage() method

Answer: D

QUESTION: 66
How can you pass values from one simulated user to another during a test?

A. create a global variable for each value in the test schedule and add references to both
store and retrieve the test values in the global variables

19 http://www.hotcerts.com
000-647

B. add custom code classes to the test using the IDataArea put() and get() methods to both
store and retrieve the test values
C. create a shared class for all tests and then use the IDataArea put() and get() methods to
both store and retrieve the test values
D. add custom code classes to the test using any Java file I/O method to write and read the
values to a persistent file

Answer: B

QUESTION: 67
What must you do if you want to reuse RPT tests on a server other than the one on which
they were recorded?

A. use the Test Search feature to find and replace the server host name(s) in each RPT test
B. select each page for each RPT test in the test editor and change the Target host to the
new server host name
C. create a new location and deployment for the new server host name(s) and add the RPT
tests as artifacts for the deployment
D. re-record the tests in the new environment because it is not possible to reuse a recorded
test on different servers

Answer: A

QUESTION: 68
What is the fastest way to debug a test by validating data returned from a server?

A. run the test and then, in the Execution History, view the page in the Protocol Data
Browser tab and manually verify the displayed value
B. enable a content verification point (VP) for the value on the page where it is displayed
C. create a reference from the returned value and then validate the reference against a
datapool using a condition test element
D. create a reference from the returned value and then validate the reference against a
datapool using an if-then statement in custom code

Answer: B

QUESTION: 69
What is one way to modify or enhance a test programmatically?

A. add Java custom code classes to a test as well as directly to the test code (*.java)
B. add Java custom code classes to a test, but not directly to the test code (*.java)
C. directly modify the test code (*.java) by adding custom code classes to the appropriate

20 http://www.hotcerts.com
000-647

files
D. modify the default test superclass with custom code

Answer: B

QUESTION: 70
How can you verify that activity is being recorded during a test capture session?

A. view the Protocol Data Response Content tab and verify that page data are displayed
B. view the test editor and verify that pages are being created for the test
C. pause the recording and view the test Java source file to verify that there are transaction
calls in the test
D. view the Recorder Control view and verify that the Kbytes recorded is not zero and
continuously updated

Answer: D

QUESTION: 71
You want to use test data in RPT custom code that comes from a server (for example, to
validate a response or to calculate a running total). What must you do to accomplish this?

A. create a datapool value that gets its value from a test reference for the server value, then
add this to the custom code argument list
B. create a reference for the server value at the beginning of the test, then add this to the
custom code argument list
C. create a reference for the server value anywhere before/above the point in the test where
the value is returned, then add this to the custom code argument list
D. create a reference for the server value at the footer of the test, then add this to the custom
code argument list

Answer: C

QUESTION: 72
Which technique of adding loops and conditionals to an RPT test provides the greatest
flexibility and is the least limited in its capability?

A. adding the loop and condition test elements in the test editor
B. adding the loop and condition test elements in a performance schedule
C. adding loops and conditional Java statements in custom code classes in the test
D. adding loops and conditional Java statements in the recorded session (*.rec)

21 http://www.hotcerts.com
000-647

Answer: C

QUESTION: 73
What is the purpose of test data management for load testing?

A. to ensure representative and realistic tests where metrics can be gathered accurately
B. to ensure that data are not retrieved from the various cache lines in the system under test
C. to ensure that tests play back correctly after recording
D. to use different user names and passwords for each iteration

Answer: A

QUESTION: 74
What is the purpose of correlation techniques during the enhancement phase of a test?

A. to ensure that the same items that were recorded are not used during the playback
B. to ensure that dynamic responses can be accommodated in subsequent requests
C. to ensure that dynamic and unique session IDs are used for each virtual user
D. to ensure that dynamic requests are datapooled to provide different input values

Answer: B

QUESTION: 75
Session IDs being used and maintained throughout a transaction is an example of ___

A. a datapool with unique user names and passwords only


B. a datapool with unique user names, passwords, and sessions IDs
C. a combination of datapooling and correlation techniques
D. built-in auto correlation features

Answer: D

QUESTION: 76
The Test Data table for a page is used to _____.

A. view the locations of test elements


B. create substitutions for datapools
C. validate the URL encoded substitutions
D. view candidates for substitution

22 http://www.hotcerts.com
000-647

Answer: D

QUESTION: 77
What does each virtual user use during every loop when the open mode of a datapool is set to
private with no wrap (i.e. the wrap check box is unchecked)?

A. a unique row from the datapool because the test will not retrieve restart from the
beginning of the datapool when the end of the datapool is reached
B. a private view of the datapool with unique rows apportioned
C. a private view of a datapool with values that can be reused by other virtual users
D. a private view of a datapool assigned per agent machine

Answer: C

QUESTION: 78
During test execution, the datapool framework is used to vary _____.

A. user names and passwords


B. data in a transaction flow
C. the contents in responses
D. request header variables

Answer: B

QUESTION: 79
If the automatic data correlation for execution is set to efficiency and the datapools are open
in private mode, test execution will be _____.

A. faster, but datapool values will not be unique for each virtual tester
B. faster, and datapool values will be unique for each virtual tester
C. slower, but datapool access will be fast because of private views
D. slower because the correlated values are substituted from the nearest previous reference

Answer: A

QUESTION: 80
By default, an element that has been substituted with a datapool variable appears in a test as
_____.

A. light blue text on a dark blue background


B. dark blue text on a light blue background

23 http://www.hotcerts.com
000-647

C. white text on a dark green background


D. dark pink text on a pink background

Answer: C

QUESTION: 81
Which is a challenge of test data management?

A. understanding the relationship of interrelated data needed for datapools and correlation
B. using the datapool import wizard when managing large amounts of data
C. developing representative and realistic tests, while minimizing the overhead introduced
by the test tool
D. managing the datapool data for all the iterations of the virtual testers across all agents

Answer: C

QUESTION: 82
A customer recently purchased RPT. The playback tests appear to have passed in the Test
Execution History. However, upon closer inspection using the Protocol Data view, some
pages following the datapool substitution failed. What is the most likely cause of this
problem?

A. The datapool was opened in the wrong shared mode.


B. The datapool variable was substituted with the incorrect test element.
C. The datapool was not checked with a verification point.
D. The datapool value was substituted, but not fully correlated.

Answer: D

QUESTION: 83
What is caused by an overloaded system under tests?

A. too much data logging


B. response failures or timeouts
C. increased CPU usage on test agents
D. increased RPT license consumption

Answer: B

QUESTION: 84
How do you control random user activity?

24 http://www.hotcerts.com
000-647

A. set an iteration rate


B. use additional custom code
C. set the number of iterations
D. define a weight for each test item

Answer: D

QUESTION: 85
What is the maximum recommended JVM heap size?

A. 1.9 GB
B. 2 GB
C. 80% of the physical memory
D. It is different for every supported OS.

Answer: A

QUESTION: 86
Which step is NOT required to validate and debug performance schedules?

A. remove any resource bottlenecks


B. validate initial response time measurements
C. run performance schedule in multi-user mode
D. find maximum user capacity of the system under test

Answer: D

QUESTION: 87
What is the purpose of the runtime settings for Statistics?

A. to set the Execution History Log Level


B. to set the Problem Determination Log Level
C. to control the amount of data that is collected during a performance schedule run
D. to control the amount of detail that is available in the statistical data from a Performance
Test

Answer: C

QUESTION: 88
Which is NOT included in the schedule elements details?

25 http://www.hotcerts.com
000-647

A. user group size


B. user load
C. execution history
D. schedule statistics

Answer: A

QUESTION: 89
What is the correct way to achieve user pacing that emulates realistic user activity?

A. use the recorded think time


B. specify a fixed think time
C. vary the think time by a random percentage
D. increase or decrease the think time by a percentage

Answer: C

QUESTION: 90
The amount of memory consumed by the workbench JVM depends on the----------

A. transaction rate
B. OS of the test agent
C. number of virtual users
D. amount of monitored data

Answer: D

QUESTION: 91
What does weighting in a random selector allow you to control?

A. the frequency of user activities


B. the average think time within a selector
C. the number of iterations of a specific user activity
D. the total run time of a specific user activity

Answer: A

QUESTION: 92
What limits the available memory for virtual users?

26 http://www.hotcerts.com
000-647

A. the JVM maximum heap size


B. the type of OS on the test agent
C. the CPU architecture (32-bit vs. 64-bit)
D. the amount of physical memory on the test agent

Answer: A

QUESTION: 93
What can the built-in resource monitoring of RPT be used to monitor?

A. Linux hosts
B. Windows hosts
C. mainframe hosts
D. all platforms supported by the TPTP monitoring framework

Answer: B

QUESTION: 94
What do the sampling options in a schedule allow you to reduce in a large system test?

A. the CPU overhead


B. the total test duration
C. the number of virtual users
D. the required network bandwidth

Answer: A

QUESTION: 95
What simulates repeated user activity?

A. loops
B. repeated blocks
C. selectors
D. delays

Answer: A

QUESTION: 96
What are the main performance schedule elements?

27 http://www.hotcerts.com
000-647

A. loops and delays


B. user groups and tests
C. selectors and comments
D. number of users and run duration

Answer: B

QUESTION: 97
Performance Test results data can be exported from RPT in XML, HTML and CSV format.
Which view contains information that can be exported in CSV format?

A. Test Navigator
B. Performance Report
C. Performance Test Runs
D. Test Execution History

Answer: C

QUESTION: 98
What does the Performance Report: Response vs. Time Detail Graph show?

A. the average page response time per interval for each individual page vs. run duration
B. the maximum page response time per interval for each individual page vs. run duration
C. the measured page response time for each individual page vs. run duration
D. the average response time for all page elements during the run vs. run duration

Answer: A

QUESTION: 99
You need to report the total server throughput for all traffic during a test run using the
default Test Report. Which counter is best to use?

A. Performance Report: Page Throughput Page Hit Rate


B. Page Element Report: Page Element Throughput Page Element Hit Rate
C. Transaction Report: Transaction Throughput Transaction Rate per Second
D. Page Element Report: Page Element Throughput Page Element Attempt Rate

Answer: B

QUESTION: 100

28 http://www.hotcerts.com
000-647

How do you view Execution History during a run?

A. You edit the Agent location file to enable real-time transfer of Execution History data.
B. Execution History is transferred once the run completes. It cannot be viewed during a
run.
C. You select an Agent host under the appropriate schedule in the Performance Test Runs
view.
D. You set the Execution History Log Level to "Schedule Actions."

Answer: A

QUESTION: 101
Which is NOT a possible report customization?

A. rearrange items by specifying a sort order


B. insert items by adding new counters to the report
C. control the appearance of items by specifying the graph type
D. focus on items of interest by filtering data

Answer: A

QUESTION: 102
Which view is most useful for identifying the cause of failing page elements?

A. Test Execution History


B. Verification Points Report: Summary
C. Page Element Report: Server Health Detail
D. Page Element Report: Response vs. Time Detail

Answer: A

QUESTION: 103
While running a performance test, the Server Health Detail report shows that pages are
failing as the load increases. You stop the run and examine the Test Execution History. It
also shows that pages are failing, but the page response Protocol Data view is blank. What
should you do?

A. examine the Test Agent error logs for the failed page error data
B. change the Execution History Log Level to "All" and re-run the test
C. leave everything the same and re-run the test, but this time do not stop it mid-way
D. change the Execution History Log Level to Secondary Test Actions and re-run the test

29 http://www.hotcerts.com
000-647

Answer: B

QUESTION: 104
It is common to exclude collected test data during ramp up and ramp down of a performance
test. What do you do to discard the last ten (10) minutes of a test for reporting purposes?

A. set the graphs X Max property to 600


B. set the graphs X Sliding Scale property to 600
C. set the graphs X Min property to the total run time minus 600
D. set the graphs X Max property to the total run time minus 600

Answer: D

QUESTION: 105
What is the first step in workload definition?

A. clarify and document the goals of the performance-testing project


B. examine the production environment being tested
C. identify test variables and user groups
D. record a performance test to verify that the application is testable

Answer: A

QUESTION: 106
What is the recommended number of variables you should change at one time for tests that
will be compared?

A. all
B. four
C. none
D. one

Answer: D

QUESTION: 107
Which type of performance test determines the conditions under which a system fails and
how it fails?

A. load
B. longevity and endurance

30 http://www.hotcerts.com
000-647

C. stress
D. volume

Answer: C

QUESTION: 108
What documents all artifacts that define the system, users, and tasks associated with the
target workload for the performance testing effort?

A. datapool document
B. product manual
C. test schedule guide
D. workload analysis document

Answer: D

QUESTION: 109
What are the most common measurement criteria collected during performance tests?

A. execution timestamp and passed verification points


B. object recognition and verification points
C. response times and throughput
D. tester names and test user groups

Answer: C

QUESTION: 110
What is the goal of the workload analysis?

A. to clarify the objective of the workload interval


B. to ensure that tests performed represent real user activity
C. to gather statistics about the system under test
D. to identify the critical measurement points for each test

Answer: B

QUESTION: 111
How can you pass values from one simulated user to another during a test?

A. create a global variable for each value in the test schedule and add references to both

31 http://www.hotcerts.com
000-647

store and retrieve the test values in the global variables


B. add custom code classes to the test using the IDataArea put() and get() methods to both
store and retrieve the test values
C. create a shared class for all tests and then use the IDataArea put() and get() methods to
both store and retrieve the test values
D. add custom code classes to the test using any Java file I/O method to write and read the
values to a persistent file

Answer: B

QUESTION: 112
What is the fastest way to debug a test by validating data returned from a server?

A. run the test and then, in the Execution History, view the page in the Protocol Data
Browser tab and manually verify the displayed value
B. enable a content verification point (VP) for the value on the page where it is displayed
C. create a reference from the returned value and then validate the reference against a
datapool using a condition test element
D. create a reference from the returned value and then validate the reference against a
datapool using an if-then statement in custom code

Answer: B

QUESTION: 113
You want to use test data in RPT custom code that comes from a server (for example, to
validate a response or to calculate a running total). What must you do to accomplish this?

A. create a datapool value that gets its value from a test reference for the server value, then
add this to the custom code argument list
B. create a reference for the server value at the beginning of the test, then add this to the
custom code argument list
C. create a reference for the server value anywhere before/above the point in the test where
the value is returned, then add this to the custom code argument list
D. create a reference for the server value at the footer of the test, then add this to the custom
code argument list

Answer: C

QUESTION: 114
You have an RPT test that does not indicate the correct number of pages in the actual web-
based application. An application goes through eight distinct pages but the RPT recorded test
only shows six pages. How can you correct this?

32 http://www.hotcerts.com
000-647

A. change the Performance Test Generation - Protocol preferences and re-generate the test
from the recorded session (.recmodel file)
B. change the HTTP Recorder Options and re-record the test
C. manually add the missing pages or remove the extra/redundant pages from the test editor
D. re-record the test and, while recording, click on New Page from the Recorder Control tab
where the pages were skipped

Answer: A

QUESTION: 115
What must you do if you want to record tests on a web browser other than Microsoft Internet
Explorer?

A. download and install the RPT IE plug-in for the supported web browser from
developerWorks
B. select Enable Environments for Testing from the Configure menu and add the web
browser as described in the product documentation
C. manually configure the browser to communicate to the server(s) via a SOCKS proxy as
described in the product documentation
D. manually configure the web browser to use the Microsoft JVM as described in the
product documentation

Answer: C

QUESTION: 116
How can you make RPT tests more readable for others?

A. add comments during recording and enter Javadoc information for each test
B. label all blocks, add comments and use a naming convention with folders
C. label all timers, rename page titles and use the recorded think time
D. rename page titles, add comments, and use a naming convention with folders

Answer: D

QUESTION: 117
What is one way to debug an RPT test by logging test values (server data, datapool
selections, etc.)?

A. add Log Entry test elements to the test in the test editor
B. use the Log Entry script support function while recording the test
C. add the test execution services reportMessage() method to the test source code

33 http://www.hotcerts.com
000-647

D. add custom code classes to the test using the reportMessage() method

Answer: D

QUESTION: 118
What is one way to modify or enhance a test programmatically?

A. add Java custom code classes to a test as well as directly to the test code (*.java)
B. add Java custom code classes to a test, but not directly to the test code (*.java)
C. directly modify the test code (*.java) by adding custom code classes to the appropriate
files
D. modify the default test superclass with custom code

Answer: B

QUESTION: 119
How can you create a copy of an RPT test

A. right-click the test; select Copy from the menu; enter a new name for the copied test
B. select the test; from the File menu, click ? Save As ; enter a new name for the copied test
C. switch to the Navigator view; select the source Java file (*.rec) of the test; from the File
menu, click ? Save As; enter a new name for the copied test
D. from the Navigator view, right-click the .recmodel file; select Generate Test; choose
Performance Test Generator; enter a new name for the copied test

Answer: D

QUESTION: 120
Which technique of adding loops and conditionals to an RPT test provides the greatest
flexibility and is the least limited in its capability?

A. adding the loop and condition test elements in the test editor
B. adding the loop and condition test elements in a performance schedule
C. adding loops and conditional Java statements in custom code classes in the test
D. adding loops and conditional Java statements in the recorded session (*.rec)

Answer: C

QUESTION: 121
What must you do to change the connection security in an existing test from SSL to non-
secure (assuming the servers are reconfigured for this)?

34 http://www.hotcerts.com
000-647

A. use the Test Search feature to find and replace each connection string from http to https
B. create a new location for each server using an SSL connection and deselect the Enable
SSL checkbox in the server details window
C. go to a performance test schedule and deselect the Enable SSL checkbox in the Schedule
Element Details connection settings
D. go to the test editor and deselect the Enable SSL checkbox in the Connection Details
Authentication and Security pane

Answer: D

QUESTION: 122
To validate correct test playback, you can enable Page Title Verification Points ---------

A. only if you have not renamed the page titles from their original recorded values
B. regardless of whether or not the page titles have been renamed
C. regardless of whether or not the page titles have been renamed, but only if each page title
has a unique name
D. regardless of whether or not the page titles have been renamed, but only if the page titles
match the expected page title values

Answer: B

QUESTION: 123
What must you do if you want to reuse RPT tests on a server other than the one on which
they were recorded?

A. use the Test Search feature to find and replace the server host name(s) in each RPT test
B. select each page for each RPT test in the test editor and change the Target host to the
new server host name
C. create a new location and deployment for the new server host name(s) and add the RPT
tests as artifacts for the deployment
D. re-record the tests in the new environment because it is not possible to reuse a recorded
test on different servers

Answer: A

QUESTION: 124
You manipulate the execution of steps in a test by repeating sequences of a transaction. How
can you add further control?

35 http://www.hotcerts.com
000-647

A. add a Loop test element and then Events and Dependencies in the test suite
B. add do or for loop statements to the test source code and then use either the breakLoop()
or the continueLoop() method
C. add a Loop test element and then custom code classes with either the breakLoop() or the
continueLoop() method
D. add a Loop test element and then Loop Control test elements in the test editor

Answer: C

QUESTION: 125
By default, an element that has been substituted with a datapool variable appears in a test as
_____.

A. light blue text on a dark blue background


B. dark blue text on a light blue background
C. white text on a dark green background
D. dark pink text on a pink background

Answer: C

QUESTION: 126
Generally, applications that use client side logic require the relevant subsequent responses to
be correlated with a _____.

A. field reference
B. reference
C. variable from a datapool
D. custom code class

Answer: D

QUESTION: 127
What is the purpose of correlation techniques during the enhancement phase of a test?

A. to ensure that the same items that were recorded are not used during the playback
B. to ensure that dynamic responses can be accommodated in subsequent requests
C. to ensure that dynamic and unique session IDs are used for each virtual user
D. to ensure that dynamic requests are datapooled to provide different input values

36 http://www.hotcerts.com
000-647

Answer: B

QUESTION: 128
Once a reference has been established, what is used to most efficiently find all subsequent
potential substitutions?

A. the Search and Replace tool


B. the Test Search tool searching all Requests
C. the "find more" and "go to" functions from the right-click submenu on a candidate
D. the Test Data preview pane under the test elements

Answer: B

QUESTION: 129
What is the purpose of test data management for load testing?

A. to ensure representative and realistic tests where metrics can be gathered accurately
B. to ensure that data are not retrieved from the various cache lines in the system under test
C. to ensure that tests play back correctly after recording
D. to use different user names and passwords for each iteration

Answer: A

QUESTION: 130
The Test Data table for a page is used to _____.

A. view the locations of test elements


B. create substitutions for datapools
C. validate the URL encoded substitutions
D. view candidates for substitution

Answer: D

QUESTION: 131
What does each virtual user use during every loop when the open mode of a datapool is set
to private with no wrap (i.e. the wrap check box is unchecked)?

A. a unique row from the datapool because the test will not retrieve restart from the
beginning of the datapool when the end of the datapool is reached
B. a private view of the datapool with unique rows apportioned
C. a private view of a datapool with values that can be reused by other virtual users

37 http://www.hotcerts.com
000-647

D. a private view of a datapool assigned per agent machine

Answer: C

QUESTION: 132
Session IDs being used and maintained throughout a transaction is an example of _____.

A. a datapool with unique user names and passwords only


B. a datapool with unique user names, passwords, and sessions IDs
C. a combination of datapooling and correlation techniques
D. built-in auto correlation features

Answer: D

QUESTION: 133
A customer recently purchased RPT. The playback tests appear to have passed in the Test
Execution History. However, upon closer inspection using the Protocol Data view, some
pages following the datapool substitution failed. What is the most likely cause of this
problem?

A. The datapool was opened in the wrong shared mode.


B. The datapool variable was substituted with the incorrect test element.
C. The datapool was not checked with a verification point.
D. The datapool value was substituted, but not fully correlated.

Answer: D

QUESTION: 134
Which is a challenge of test data management?

A. understanding the relationship of interrelated data needed for datapools and correlation
B. using the datapool import wizard when managing large amounts of data
C. developing representative and realistic tests, while minimizing the overhead introduced
by the test tool
D. managing the datapool data for all the iterations of the virtual testers across all agents

Answer: C

QUESTION: 135
What limits the available memory for virtual users?

38 http://www.hotcerts.com
000-647

A. the JVM maximum heap size


B. the type of OS on the test agent
C. the CPU architecture (32-bit vs. 64-bit)
D. the amount of physical memory on the test agent

Answer: A

QUESTION: 136
The level of detail in the Execution History is dependent on _____.

A. log level
B. group size setting
C. the user think time limit
D. the number of users

Answer: A

QUESTION: 137
What simulates repeated user activity?

A. loops
B. repeated blocks
C. selectors
D. delays

Answer: A

QUESTION: 138
What is the correct way to achieve user pacing that emulates realistic user activity?

A. use the recorded think time


B. specify a fixed think time
C. vary the think time by a random percentage
D. increase or decrease the think time by a percentage

Answer: C

QUESTION: 139
Which is NOT included in the schedule elements details?

39 http://www.hotcerts.com
000-647

A. user group size


B. user load
C. execution history
D. schedule statistics

Answer: A

QUESTION: 140
What is caused by an overloaded system under tests?

A. too much data logging


B. response failures or timeouts
C. increased CPU usage on test agents
D. increased RPT license consumption

Answer: B

QUESTION: 141
What is the maximum recommended JVM heap size?

A. 1.9 GB
B. 2 GB
C. 80% of the physical memory
D. It is different for every supported OS.

Answer: A

QUESTION: 142
Which step is NOT required to validate and debug performance schedules?

A. remove any resource bottlenecks


B. validate initial response time measurements
C. run performance schedule in multi-user mode
D. find maximum user capacity of the system under test

Answer: D

QUESTION: 143
What is the purpose of the runtime settings for Statistics?

40 http://www.hotcerts.com
000-647

A. to set the Execution History Log Level


B. to set the Problem Determination Log Level
C. to control the amount of data that is collected during a performance schedule run
D. to control the amount of detail that is available in the statistical data from a Performance
Test

Answer: C

QUESTION: 144
The amount of memory consumed by the workbench JVM depends on the _____.

A. transaction rate
B. OS of the test agent
C. number of virtual users
D. amount of monitored data

Answer: D

QUESTION: 145
How do you control random user activity?

A. set an iteration rate


B. use additional custom code
C. set the number of iterations
D. define a weight for each test item

Answer: D

QUESTION: 146
What do the sampling options in a schedule allow you to reduce in a large system test?

A. the CPU overhead


B. the total test duration
C. the number of virtual users
D. the required network bandwidth

Answer: A

QUESTION: 147
What is used to model a web page hit rate?

41 http://www.hotcerts.com
000-647

A. loops
B. delays
C. paced loops
D. random selectors

Answer: C

QUESTION: 148
What are the main performance schedule elements?

A. loops and delays


B. user groups and tests
C. selectors and comments
D. number of users and run duration

Answer: B

QUESTION: 149
Which view is most useful for identifying the cause of failing page elements?

A. Test Execution History


B. Verification Points Report: Summary
C. Page Element Report: Server Health Detail
D. Page Element Report: Response vs. Time Detail

Answer: A

QUESTION: 150
While running a performance test, the Server Health Detail report shows that pages are
failing as the load increases. You stop the run and examine the Test Execution History. It
also shows that pages are failing, but the page response Protocol Data view is blank. What
should you do?

A. examine the Test Agent error logs for the failed page error data
B. change the Execution History Log Level to "All" and re-run the test
C. leave everything the same and re-run the test, but this time do not stop it mid-way
D. change the Execution History Log Level to Secondary Test Actions and re-run the test

Answer: B
QUESTION: 151

42 http://www.hotcerts.com
000-647

How do you view Execution History during a run?

A. You edit the Agent location file to enable real-time transfer of Execution History data.
B. Execution History is transferred once the run completes. It cannot be viewed during a
run.
C. You select an Agent host under the appropriate schedule in the Performance Test Runs
view.
D. You set the Execution History Log Level to "Schedule Actions."

Answer: A

QUESTION: 152
What does the Performance Report: Response vs. Time Detail Graph show?

A. the average page response time per interval for each individual page vs. run duration
B. the maximum page response time per interval for each individual page vs. run duration
C. the measured page response time for each individual page vs. run duration
D. the average response time for all page elements during the run vs. run duration

Answer: A

QUESTION: 153
Which is NOT a possible report customization?

A. rearrange items by specifying a sort order


B. insert items by adding new counters to the report
C. control the appearance of items by specifying the graph type
D. focus on items of interest by filtering data

Answer: A

QUESTION: 154
It is common to exclude collected test data during ramp up and ramp down of a performance
test. What do you do to discard the last ten (10) minutes of a test for reporting purposes?

A. set the graphs X Max property to 600


B. set the graphs X Sliding Scale property to 600
C. set the graphs X Min property to the total run time minus 600
D. set the graphs X Max property to the total run time minus 600

43 http://www.hotcerts.com
000-647

Answer: D

QUESTION: 155
You need to report the total server throughput for all traffic during a test run using the
default Test Report. Which counter is best to use?

A. Performance Report: Page Throughput Page Hit Rate


B. Page Element Report: Page Element Throughput Page Element Hit Rate
C. Transaction Report: Transaction Throughput Transaction Rate per Second
D. Page Element Report: Page Element Throughput Page Element Attempt Rate

Answer: B

QUESTION: 156
Performance Test results data can be exported from RPT in XML, HTMLand CSV format.
Which view contains information that can be exported in CSV format?

A. Test Navigator
B. Performance Report
C. Performance Test Runs
D. Test Execution History

Answer: C

44 http://www.hotcerts.com

You might also like