Monitor Publish performance during load testing in Adobe Experience Manager as a Cloud Service
During a load test on AEM as a Cloud Service Publish, use the APM integration enabled for your program to review application performance, including response time, errors, and throughput. These metrics help assess the user-visible effect of increased traffic; they do not, by themselves, confirm when a publisher pod was created or became ready. To validate the test, compare application performance before, during, and after the load period, and record any errors or latency changes.
Description description
Environment
- Adobe Experience Manager as a Cloud Service Publish environment
- The APM integration available to the program, such as New Relic One or OpenTelemetry where enabled
Issue/Symptoms
- You are load testing the Publish tier and need to assess the effect of increased traffic on application response time, errors, and throughput.
- You want to know whether the APM data confirms a specific publisher scale-out event or pod-ready time.
Cause
AEM as a Cloud Service manages Publish-tier scaling as part of the platform. Customer APM provides application observability; it does not provide a change tracker for scale-up pipelines. New Relic’s documented change trackers cover supported deployment and update workflows, and explicitly exclude scale-up pipelines. Use APM metrics to assess performance during the test, not as proof of the exact time a publisher pod was added.
Adobe is replacing the built-in New Relic integration with OpenTelemetry, which is in Limited Availability. Check which APM integration is enabled for your environment before following monitoring instructions.
Resolution resolution
Follow the steps below to resolve the issue:
- Monitor auto-scaling events during load testing and observe end-user response times and error rates to identify when auto-scaling occurs and when new Publisher instances become available.
- Use Application Performance Monitoring (APM) tools, such as New Relic One, to validate the transition during scaling events.
- Monitor response times, error rates, JVM performance, transaction times, and external or database calls, and verify that no significant spikes occur during scaling events.
- Review performance metrics, including CPU utilization, memory usage, and transaction data, in New Relic One.
- If you do not have access to New Relic One, contact your organization’s AEM administrator or support contact to request access or activation.
- If additional monitoring or custom dashboards are required, refer to the official AEM documentation or support resources for guidance.
- Verify that you can observe real-time performance metrics and auto-scaling events during load testing and confirm that scaling transitions occur without significant impacts on response times or error rates.