Load testing a Pega application 

To load test your Pega application, you can use any web application load-testing tool, such as jMeter or Loadrunner.

Before exercising a performance test, the best practice is to exercise the main paths through the application, including all those to be exercised by the test scripts, and then take a Performance Analyzer (PAL) reading for each path. Investigate and fix any issues that are exposed.

Run the load test as a series of iterations with the goals identified by business metrics along with technical metrics to achieve.

• Test environment baseline – This is the first test to establish that application, environment, and tools are all working correctly.
• Application baseline – This is a test run with one user or one batch process creating a case in a single JVM. Then increase to 10 then to 100 users or 100 cases created by the batch process.
• Full end to end test – This is the first full test of the application end to end, still in a single JVM.
• Failure in one JVM – Test what happens if there is a failure in one of the JVMs.
• Span JVMs based on the peak business and technical metrics/goals – This is iterated as much as is needed to achieve agreed success metrics.

Begin testing just with HTTP transactions by disabling agents and listeners. Then, test the agents and listeners. Finally, test with both foreground and background processing.

The performance tests must be designed to mimic the real-world production use. Collect data on CPU utilization, I/O volume, memory utilization, and network utilization to help understand the influences on performance.

Relate the capacity of the test machines to production hardware. If the test machines have 20 percent of the performance of the production machines, then the test workload should be 20 percent of the expected production workload. If you expect to use two or more JVMs per server in production, use the same number when testing.