Testing for concurrency | Scalability testing


Analyzing stress test results


For each test scenario executed:

If any of these passing criterion are not met then it should be considered a defect. If error rate is the cause of the test case failure then the defective code needs investigation.

If overall low performance is a concern then we need to go through the iterative process of performance tuning and re-executing the test case, as shown in Figure 23-1.

Last but not the least, if the concern is memory leak, core dump, or gradual throughput degradation, then refer to Chapter 24, Analyzing test results and solving performance problems.