解释JMeter报告结果

Question

Can you please explain why throughput is almost the same but number of samples is much lower in run with 400 threads/users? Error% is same as well. Application under test is java, cpu load in both cases/runs is 90%

Answer 1

How do both tests durations correlate? Throughput is being calculated like:

Throughput = (number of requests) / (total time)

So my expectation is that your "400 users" test duration is ~2.5x times higher than for "150" users test.

References:

• JMeter Glossary
• Understanding Your Reports: Part 2 - KPI Correlations