feat: store benchmarks in s3 #2668
Workflow file for this run
This file contains 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: PR Benchmarks | |
| on: | |
| pull_request: | |
| types: [ labeled, synchronize ] | |
| branches: [ "develop" ] | |
| workflow_dispatch: { } | |
| permissions: | |
| actions: write | |
| contents: read | |
| pull-requests: write | |
| id-token: write | |
| jobs: | |
| label_trigger: | |
| runs-on: ubuntu-latest | |
| if: ${{ contains(github.event.head_commit.message, '[benchmark]') || github.event.label.name == 'benchmark' && github.event_name == 'pull_request' }} | |
| steps: | |
| # We remove the benchmark label first so that the workflow can be re-triggered. | |
| - uses: actions-ecosystem/action-remove-labels@v1 | |
| with: | |
| labels: benchmark | |
| bench: | |
| needs: label_trigger | |
| strategy: | |
| matrix: | |
| benchmark: | |
| - id: datafusion | |
| name: DataFusion | |
| - id: random_access | |
| name: Random Access | |
| - id: compress | |
| name: Vortex Compression | |
| runs-on: self-hosted | |
| if: ${{ contains(github.event.head_commit.message, '[benchmark]') || github.event.label.name == 'benchmark' && github.event_name == 'pull_request' }} | |
| steps: | |
| - uses: actions/checkout@v4 | |
| - uses: ./.github/actions/cleanup | |
| - uses: ./.github/actions/setup-rust | |
| - uses: spiraldb/actions/.github/actions/[email protected] | |
| # The compression benchmarks rely on DuckDB being installed to convert CSV to Parquet | |
| - name: Install DuckDB | |
| uses: opt-nc/[email protected] | |
| if: runner.environment != 'self-hosted' | |
| with: | |
| version: v1.0.0 | |
| - name: Set tempdir | |
| if: runner.environment == 'self-hosted' | |
| run: | | |
| echo "TMPDIR=/work" >> $GITHUB_ENV | |
| - name: Run benchmark | |
| shell: bash | |
| env: | |
| BENCH_VORTEX_RATIOS: '.*' | |
| RUSTFLAGS: '-C target-cpu=native' | |
| run: | | |
| cargo install cargo-criterion | |
| sudo apt-get update && sudo apt-get install -y jq | |
| cargo criterion \ | |
| --bench ${{ matrix.benchmark.id }} \ | |
| --message-format=json \ | |
| > ${{ matrix.benchmark.id }}-raw.json | |
| cat ${{ matrix.benchmark.id }}-raw.json \ | |
| | bash scripts/coerce-criterion-json.sh \ | |
| > ${{ matrix.benchmark.id }}.json | |
| - name: Setup AWS CLI | |
| uses: aws-actions/configure-aws-credentials@v1 | |
| with: | |
| role-to-assume: arn:aws:iam::375504701696:role/GitHubBenchmarkRole | |
| aws-region: us-east-1 | |
| - name: Compare results | |
| shell: bash | |
| run: | | |
| set -Eeu -o pipefail -x | |
| base_commit_sha=${{ github.event.pull_request.base.sha }} | |
| aws s3 cp s3://vortex-benchmark-results-database/data.json - \ | |
| | grep $base_commit_sha \ | |
| > base.json | |
| echo '# Benchmarks: ${{ matrix.benchmark.id }}' > comment.md | |
| uv run pip install pandas | |
| uv run python3 scripts/compare-benchmark-jsons.py base.json ${{ matrix.benchmark.id }}.json \ | |
| >> comment.md | |
| - name: Comment PR | |
| uses: thollander/actions-comment-pull-request@v3 | |
| with: | |
| file-path: comment.md | |
| comment-tag: execution | |
| tpch: | |
| needs: label_trigger | |
| runs-on: self-hosted | |
| steps: | |
| - uses: actions/checkout@v4 | |
| - uses: ./.github/actions/cleanup | |
| - uses: ./.github/actions/setup-rust | |
| - uses: spiraldb/actions/.github/actions/[email protected] | |
| # The compression benchmarks rely on DuckDB being installed to convert CSV to Parquet | |
| - name: Install DuckDB | |
| uses: opt-nc/[email protected] | |
| if: runner.environment != 'self-hosted' | |
| with: | |
| version: v1.0.0 | |
| - name: Set tempdir | |
| if: runner.environment == 'self-hosted' | |
| run: | | |
| echo "TMPDIR=/work" >> $GITHUB_ENV | |
| - name: Run TPC-H benchmark | |
| shell: bash | |
| env: | |
| BENCH_VORTEX_RATIOS: '.*' | |
| RUSTFLAGS: '-C target-cpu=native' | |
| run: | | |
| cargo run --bin tpch_benchmark --release -- -d gh-json -t 1 | tee tpch.json | |
| - name: Setup AWS CLI | |
| uses: aws-actions/configure-aws-credentials@v1 | |
| with: | |
| role-to-assume: arn:aws:iam::375504701696:role/GitHubBenchmarkRole | |
| aws-region: us-east-1 | |
| - name: Compare results | |
| shell: bash | |
| run: | | |
| set -Eeu -o pipefail -x | |
| base_commit_sha=${{ github.event.pull_request.base.sha }} | |
| aws s3 cp s3://vortex-benchmark-results-database/data.json - \ | |
| | grep $base_commit_sha \ | |
| > base.json | |
| echo '# Benchmarks: TPC-H' > comment.md | |
| uv run pip install pandas | |
| uv run python3 scripts/compare-benchmark-jsons.py base.json tpch.json \ | |
| >> comment.md | |
| - name: Comment PR | |
| uses: thollander/actions-comment-pull-request@v3 | |
| with: | |
| file-path: comment.md | |
| comment-tag: execution | |
| clickbench: | |
| needs: label_trigger | |
| runs-on: self-hosted | |
| steps: | |
| - uses: actions/checkout@v4 | |
| - uses: ./.github/actions/cleanup | |
| - uses: ./.github/actions/setup-rust | |
| - uses: spiraldb/actions/.github/actions/[email protected] | |
| # The compression benchmarks rely on DuckDB being installed to convert CSV to Parquet | |
| - name: Install DuckDB | |
| uses: opt-nc/[email protected] | |
| if: runner.environment != 'self-hosted' | |
| with: | |
| version: v1.0.0 | |
| - name: Set tempdir | |
| if: runner.environment == 'self-hosted' | |
| run: | | |
| echo "TMPDIR=/work" >> $GITHUB_ENV | |
| - name: Run Clickbench benchmark | |
| shell: bash | |
| env: | |
| BENCH_VORTEX_RATIOS: '.*' | |
| RUSTFLAGS: '-C target-cpu=native' | |
| HOME: /home/ci-runner | |
| run: | | |
| cargo run --bin clickbench --release -- -d gh-json | tee clickbench.json | |
| - name: Setup AWS CLI | |
| uses: aws-actions/configure-aws-credentials@v1 | |
| with: | |
| role-to-assume: arn:aws:iam::375504701696:role/GitHubBenchmarkRole | |
| aws-region: us-east-1 | |
| - name: Compare results | |
| shell: bash | |
| run: | | |
| set -Eeu -o pipefail -x | |
| base_commit_sha=${{ github.event.pull_request.base.sha }} | |
| aws s3 cp s3://vortex-benchmark-results-database/data.json - \ | |
| | grep $base_commit_sha \ | |
| > base.json | |
| echo '# Benchmarks: Clickbench' > comment.md | |
| uv run pip install pandas | |
| uv run python3 scripts/compare-benchmark-jsons.py base.json clickbench.json \ | |
| >> comment.md | |
| - name: Comment PR | |
| uses: thollander/actions-comment-pull-request@v3 | |
| with: | |
| file-path: comment.md | |
| comment-tag: execution |