Epic Goal

Ensure the Builds for OpenShift can handle high workloads and concurrent operations with acceptable resource usage and response time under sustained load.

Why is this important?

This is important to get the product benchmark and handle the limitations.

Test Scenarios

Concurrency Stress Test
Submit hundreds of builds in parallel, measure throughput & success rate
Duration Tests
Run continuous builds for 24h to observe resource leaks or performance degradation
Scaling Load
Gradually increase number of builds and namespaces
Operator Under Pressure
Restart builds operator under high load and test reconciliation time
Mixed Workload Test
Combine short/fast and long/heavy builds
System Degradation (optional)
Introduce degraded nodes or reduced quota to observe behavior

With those Scenarios we would like to measure:
Build Lantency:
Time from Build creation to Build start
Time from Build start to completion
Throughput:
Builds completed per unit of time (e.g., per minute/hour)
Resource Utilization:
CPU and Memory usage of builds controller/manager pods
Cluster-wide CPU/mem/disk under build load
Operator Performance:
Reaction time to config changes and events

Dependencies (internal and external)

1. ...
   We will need the performance and scale team to help us with those scenarios

Open questions::

1. …

Done Checklist

TBD

Shipwright metrics:
https://shipwright.io/docs/build/metrics/