Dear Navjot,
thank you for reaching out with your question about memory spikes in Azure Load Testing engines during execution. Let me explain this in simple terms and provide some guidance to help resolve the issue.
First, high memory usage in Azure Load Testing engines can happen when the test generates a large number of requests or processes complex responses. Each engine handles a portion of the load, and if the test requires significant memory per user, the engines may struggle even with increased counts.
Reduce the test concurrency per engine: Even if you increase the number of engines, each engine may still be handling too many users. Try lowering the number of concurrent users per engine and observe memory usage. Microsoft’s documentation suggests adjusting the user distribution for better performance. You can find more details here: Azure Load Testing documentation.
Review test script efficiency: If your test scripts include large payloads, complex logic, or store too much data in memory, this can cause spikes. Simplifying scripts or breaking them into smaller steps may help.
Monitor dependencies and response sizes: If your test interacts with services that return large responses, the engines must process and store this data, increasing memory use. Check if responses can be filtered or if unnecessary data can be ignored.
Engine configuration and limits: Azure Load Testing engines have default resource allocations. If your test exceeds these, consider reaching out to Microsoft Support to discuss possible adjustments.
Since CPU usage is normal, the issue is likely memory-specific. Try these steps and monitor changes.
I hope this helps....
Best regards,
Alex
P.S. If my answer help to you, please Accept my answer
PPS That is my Answer and not a Comment
https://ctrlaltdel.blog/