benchmark-ab #3
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| name: benchmark-ab | |
| # Manual A/B throughput comparison between two refs. | |
| # | |
| # This is a measurement, not a test: it never fails the build on a slow result. | |
| # Absolute req/s from a shared runner is meaningless on its own, so both refs | |
| # are built and measured alternately in the same job and only the ratio of the | |
| # medians is reported, with a permutation test to say whether the difference | |
| # stands out from the run-to-run noise. | |
| # | |
| # Non-SSL and Linux only for now. | |
| on: | |
| workflow_dispatch: | |
| inputs: | |
| base: | |
| description: "Baseline ref" | |
| required: false | |
| default: "origin/master" | |
| head: | |
| description: "Ref to compare (defaults to the ref this run was started on)" | |
| required: false | |
| default: "" | |
| rounds: | |
| description: "Measurement rounds per ref (9+ recommended; below 4 the test can never reach significance)" | |
| required: false | |
| default: "9" | |
| duration: | |
| description: "Load duration per measurement" | |
| required: false | |
| default: "5s" | |
| connections: | |
| description: "Concurrent connections" | |
| required: false | |
| default: "10" | |
| permissions: | |
| contents: read | |
| jobs: | |
| ubuntu: | |
| runs-on: ubuntu-latest | |
| steps: | |
| - name: checkout | |
| uses: actions/checkout@v4 | |
| with: | |
| fetch-depth: 0 | |
| - name: install bombardier | |
| run: go install github.com/codesenberg/bombardier@latest | |
| - name: run A/B benchmark | |
| run: | | |
| export PATH="$(go env GOPATH)/bin:$PATH" | |
| HEAD_REF="${{ inputs.head }}" | |
| if [ -z "$HEAD_REF" ]; then HEAD_REF="${{ github.sha }}"; fi | |
| ./benchmark/ab.sh \ | |
| --base "${{ inputs.base }}" \ | |
| --head "$HEAD_REF" \ | |
| --rounds "${{ inputs.rounds }}" \ | |
| --duration "${{ inputs.duration }}" \ | |
| --connections "${{ inputs.connections }}" |