Use application performance metrics

You can configure your workloads on Google Kubernetes Engine (GKE) to sendapplication performance metrics to Cloud Monitoring. You can thenuse these metrics to detect performance regressions in your application.Google Kubernetes Engine provides visualizations for the following kinds of performancemeasures for your workloads:

  • Requests: shows the per-second request rate, grouped by operation whenavailable.
  • Errors: shows error rates, grouped by operation and response code.
  • Latency: shows 50th and 95th percentile response latency by operation.
  • CPU and memory: shows the utilization of CPU and memory as a percentage of arequested amount.

These metrics correspond to thegolden signalsrecommended in the GoogleSite Reliability Engineering bookfor monitoring distributed systems.

You can also view and explore logs for your workloads.

Before you can use application performance metrics, your application must havea way to send the metrics to Cloud Monitoring. For information aboutrecommended approaches, seeCollect application performance metrics.

Collect application performance metrics

You can collect application performance metrics for Google Kubernetes Engine by usingthe following supported integrations:

  • Cloud Service Mesh: If you use Cloud Service Mesh, then applicationperformance metrics are collected automatically.
  • Istio: If you useopen source Istio, then werecommend that youcollect the metricsby using Google Cloud Managed Service for Prometheus.
  • GKE Ingress: When you configureGKE Ingress for Application Load Balancers,performance metrics are automatically collected for the HTTP/S load balancersthat route traffic to your GKE Service and Deployment resourcesbehind GKE Ingress.
  • NGINX Ingress: If you are usingNGINX Ingress, thenwe recommend that youcollect the metricsby using Google Cloud Managed Service for Prometheus.
  • Prometheus HTTP and gRPC metrics: If your application exposes PrometheusHTTP or gRPC metrics, then we recommend that you follow theHTTP server andgRPC serverdocuments to collect the metrics by using Google Cloud Managed Service for Prometheus.

View application performance metrics

After the performance data is available for analysis, you canview the metricsfor a Deployment on the application dashboard.

Except as otherwise noted, the content of this page is licensed under theCreative Commons Attribution 4.0 License, and code samples are licensed under theApache 2.0 License. For details, see theGoogle Developers Site Policies. Java is a registered trademark of Oracle and/or its affiliates.

Last updated 2026-02-18 UTC.