Automated benchmark test execution in CI pipelines is currently driven by actual production request data. An example of this would be a request containing various dynamic fields, such as timestamp, request ID, and token, causing the execution result of the same benchmark test to vary with every execution.
I want to gain an understanding of how these dynamic values are usually processed during automated testing (for example, normalisation or obfuscation) to ensure they are still consistent and reflect live production performance, yet allow the benchmark tests to remain deterministic and repeatable.
I want to gain an understanding of how these dynamic values are usually processed during automated testing (for example, normalisation or obfuscation) to ensure they are still consistent and reflect live production performance, yet allow the benchmark tests to remain deterministic and repeatable.