<< Identifying Performance Testing Requirements | Implementing Performance Tests >>

Designing a Realistic Workload

Planning Performance Tests

253

It is best to change only one variable at a time. Then, if performance changes with
the next test, you know that the change was caused by that one variable.

You set test variables when you set up a suite. For more information, see
Implementing Tests as Suites on page 95.

Virtual tester classifications

You categorize the virtual testers into groups based

on the types of activities they perform.

For each group, identify the number of virtual testers or the percentage of overall
virtual testers. For example, you could group 20% of the virtual testers into
Accounting, 30% of the virtual testers into Data Entry, and 50% of the virtual
testers into Sales.

Set up user groups within a suite. Keep these user groups in mind as you plan the
test scripts to be associated with the test. The test scripts should accurately reflect
the actions of realistic user groups. For information about setting up user groups,
see Inserting User Groups into a Suite on page 265.

User work profiles

The set of activities that the virtual testers perform and the

frequency with which they perform them. The virtual tester actions should mirror
the mix of tasks that the users actually perform as closely as possible.

For example, if the Sales user group accesses the database 70% more than the other
two groups, be sure that the workload reflects this.

User characteristics

Determine how long a virtual tester pauses before executing

a transaction, the rates at which the transaction is executed, and the number of
times a transaction is executed consecutively. It is important to model the real user
characteristics accurately because the values directly affect the overall
performance of the system.

For example, a user who thinks for 5 seconds and types 30 words per minute puts
a much smaller workload on the system than a user who thinks for 1 second and
types 60 words per minute.

Use delays and think times to model the virtual tester characteristics. For more
information about delays, see Inserting a Delay on page 280. For more information
about think times, see the

Rational Robot User's Guide

When designing a workload model, make sure to consider factors such as these to
ensure an accurate test environment. Taking the time to consider these factors saves
time in the long run. The more clearly defined your testing goals are, the more quickly
you can achieve them.