Documentation
¶
Overview ¶
Stress benchmark validating a multi-component snapshot with a configurable number of workers, simulating real-world release pipeline workloads. The component count and worker count are controlled via the EC_STRESS_COMPONENTS and EC_STRESS_WORKERS environment variables respectively. Uses the same golden-container image data as the simple benchmark, duplicated across components to create memory pressure. The prepare_data.sh script can be used to re-populate the data directory.
Click to show internal directories.
Click to hide internal directories.