Degrading TPS results when testing the TPC-E test (4375599)

Degrading TPS results when testing the TPC-E test

The customer is getting degraded TPS results when testing TPC-E while wanting a steady state of TPS results as per the official website of the TPC-E benchmark.

Incorrect usage of the No-Delay option instead of Real-World Delays.

BMF Dev team has tested getting a steady state with Real-World Delays of 400ms (Keying time=200ms and Think time =200ms).
To get a steady state, customer should use Real-World Delays as per their requirement.

BMF-5813_doc

Note: The issue is by using No delay in Latency. The No Delay option is used when the goal of the test is to stress the system to its limits, without concern for accurately simulating users. That’s why the Average response time increased and TPS decreased because this customer is not getting a steady state. With the No-delay option, we will not achieve a steady state. To attain a steady state, we need to use real-world delays.