Kubernetes CPU scalability
.svg)
.svg)
Description
Test that your service scales as expected when CPU capacity is limited. Gremlin will consume CPU in 3 stages: 50%, 75%, and 90%.
What this Scenario does
This Scenario progressively consumes CPU within your Kubernetes pods in three stages: 50%, 75%, and 90%. This simulates CPU pressure from traffic spikes or resource-intensive workloads, testing your HPA configuration, pod resource limits, and cluster autoscaler.
Why run this Scenario?
- Validate that the Horizontal Pod Autoscaler (HPA) triggers at the correct CPU thresholds and scales pods effectively.
- Verify that pod CPU resource requests and limits are tuned for both performance and cost efficiency.
- Test cluster autoscaler behavior when node-level CPU capacity is exhausted.
- Ensure that Kubernetes pod priorities correctly manage resources when CPU is scarce across the cluster.
Expected outcome
As CPU usage increases in Kubernetes pods, the Horizontal Pod Autoscaler (HPA) scales the workload to maintain performance within your defined thresholds.
