Skip to content

Build once, benchmark many #7289

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Draft
AdamGS wants to merge 1 commit into
base: develop
Choose a base branch
from
+149 −34
Draft

Build once, benchmark many #7289

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
54 changes: 44 additions & 10 deletions .github/workflows/bench-pr.yml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -19,7 +19,43 @@ permissions:
id-token: write # enables AWS-GitHub OIDC

jobs:
build:
timeout-minutes: 60
runs-on: >-
${{ github.repository == 'vortex-data/vortex'
&& format('runs-on={0}/runner=bench-dedicated/tag=bench-build{1}', github.run_id, github.event.pull_request.head.repo.fork == false && '/extras=s3-cache' || '')
|| 'ubuntu-latest' }}
steps:
- uses: runs-on/action@v2
if: github.event.pull_request.head.repo.fork == false
with:
sccache: s3
- uses: actions/checkout@v6
with:
ref: ${{ github.event.pull_request.head.sha }}
- uses: ./.github/actions/setup-rust
with:
repo-token: ${{ secrets.GITHUB_TOKEN }}
- name: Build binaries
shell: bash
env:
RUSTFLAGS: "-C target-cpu=native -C force-frame-pointers=yes"
run: |
cargo build --package random-access-bench --package compress-bench --profile release_debug
- name: Stage binaries for upload
shell: bash
run: |
mkdir -p staging/bin
cp target/release_debug/random-access-bench staging/bin/
cp target/release_debug/compress-bench staging/bin/
- uses: actions/upload-artifact@v7
with:
name: bench-binaries
path: staging/
retention-days: 1

bench:
needs: build
timeout-minutes: 120
runs-on: >-
${{ github.repository == 'vortex-data/vortex'
Expand All @@ -30,22 +66,16 @@ jobs:
benchmark:
- id: random-access-bench
name: Random Access
build_args: "--features lance"
- id: compress-bench
name: Compression
steps:
- uses: runs-on/action@v2
if: github.event.pull_request.head.repo.fork == false
with:
sccache: s3
- uses: actions/checkout@v6
with:
ref: ${{ github.event.pull_request.head.sha }}
- name: Setup benchmark environment
run: sudo bash scripts/setup-benchmark.sh
- uses: ./.github/actions/setup-rust
with:
repo-token: ${{ secrets.GITHUB_TOKEN }}

- name: Install DuckDB
run: |
Expand All @@ -55,12 +85,16 @@ jobs:

- uses: ./.github/actions/system-info

- name: Build binary
- uses: actions/download-artifact@v8
with:
name: bench-binaries
path: staging
- name: Install binaries
shell: bash
env:
RUSTFLAGS: "-C target-cpu=native -C force-frame-pointers=yes"
run: |
cargo build --package ${{ matrix.benchmark.id }} --profile release_debug ${{ matrix.benchmark.build_args }}
mkdir -p target/release_debug
cp staging/bin/* target/release_debug/
chmod +x target/release_debug/*

- name: Setup Polar Signals
if: github.event.pull_request.head.repo.fork == false
Expand Down
54 changes: 43 additions & 11 deletions .github/workflows/bench.yml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -31,7 +31,41 @@ jobs:
bash scripts/commit-json.sh > new-commit.json
bash scripts/cat-s3.sh vortex-ci-benchmark-results commits.json new-commit.json

build:
timeout-minutes: 60
runs-on: >-
${{ github.repository == 'vortex-data/vortex'
&& format('runs-on={0}/runner=bench-dedicated/extras=s3-cache/tag=bench-build', github.run_id)
|| 'ubuntu-latest' }}
steps:
- uses: runs-on/action@v2
if: github.repository == 'vortex-data/vortex'
with:
sccache: s3
- uses: actions/checkout@v6
- uses: ./.github/actions/setup-rust
with:
repo-token: ${{ secrets.GITHUB_TOKEN }}
- name: Build binaries
shell: bash
env:
RUSTFLAGS: "-C target-cpu=native -C force-frame-pointers=yes"
run: |
cargo build --bin random-access-bench --bin compress-bench --profile release_debug --features lance
- name: Stage binaries for upload
shell: bash
run: |
mkdir -p staging/bin
cp target/release_debug/random-access-bench staging/bin/
cp target/release_debug/compress-bench staging/bin/
- uses: actions/upload-artifact@v7
with:
name: bench-binaries
path: staging/
retention-days: 1

bench:
needs: build
timeout-minutes: 120
runs-on: >-
${{ github.repository == 'vortex-data/vortex'
Expand All @@ -42,23 +76,16 @@ jobs:
benchmark:
- id: random-access-bench
name: Random Access
build_args: "--features lance"
formats: "parquet,lance,vortex"
- id: compress-bench
name: Compression
build_args: "--features lance"
formats: "parquet,lance,vortex"
steps:
- uses: runs-on/action@v2
if: github.repository == 'vortex-data/vortex'
with:
sccache: s3
- uses: actions/checkout@v6
- name: Setup benchmark environment
run: sudo bash scripts/setup-benchmark.sh
- uses: ./.github/actions/setup-rust
with:
repo-token: ${{ secrets.GITHUB_TOKEN }}

- name: Install DuckDB
run: |
Expand All @@ -68,12 +95,16 @@ jobs:

- uses: ./.github/actions/system-info

- name: Build binary
- uses: actions/download-artifact@v8
with:
name: bench-binaries
path: staging
- name: Install binaries
shell: bash
env:
RUSTFLAGS: "-C target-cpu=native -C force-frame-pointers=yes"
run: |
cargo build --bin ${{ matrix.benchmark.id }} --profile release_debug ${{ matrix.benchmark.build_args }}
mkdir -p target/release_debug
cp staging/bin/* target/release_debug/
chmod +x target/release_debug/*

- name: Setup Polar Signals
uses: polarsignals/gh-actions-ps-profiling@v0.8.1
Expand Down Expand Up @@ -115,6 +146,7 @@ jobs:
secrets: inherit
with:
mode: "develop"
build_lance: true
benchmark_matrix: |
[
{
Expand Down
1 change: 1 addition & 0 deletions .github/workflows/nightly-bench.yml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -23,6 +23,7 @@ jobs:
with:
mode: "develop"
machine_type: ${{ matrix.machine_type.instance_name }}
build_lance: true
# datafusion:vortex uses a lot of memory
benchmark_matrix: |
[
Expand Down
74 changes: 61 additions & 13 deletions .github/workflows/sql-benchmarks.yml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -10,6 +10,10 @@ on:
required: false
type: string
default: c6id.8xlarge
build_lance:
required: false
type: boolean
default: false
benchmark_matrix:
required: false
type: string
Expand Down Expand Up @@ -98,7 +102,55 @@ on:
]

jobs:
build:
timeout-minutes: 60
runs-on: >-
${{ github.repository == 'vortex-data/vortex'
&& format('runs-on={0}/runner=bench-dedicated/tag=sql-build{1}', github.run_id, (inputs.mode != 'pr' || github.event.pull_request.head.repo.fork == false) && '/extras=s3-cache' || '')
|| 'ubuntu-latest' }}
steps:
- uses: runs-on/action@v2
if: inputs.mode != 'pr' || github.event.pull_request.head.repo.fork == false
with:
sccache: s3
- uses: actions/checkout@v6
if: inputs.mode == 'pr'
with:
ref: ${{ github.event.pull_request.head.sha }}
- uses: actions/checkout@v6
if: inputs.mode != 'pr'
- uses: ./.github/actions/setup-rust
with:
repo-token: ${{ secrets.GITHUB_TOKEN }}
- name: Build binaries
shell: bash
env:
RUSTFLAGS: "-C target-cpu=native -C force-frame-pointers=yes"
run: |
packages="--bin data-gen --bin datafusion-bench --bin duckdb-bench"
if [ "${{ inputs.build_lance }}" = "true" ]; then
packages="$packages --bin lance-bench"
fi
cargo build $packages --profile release_debug
- name: Stage binaries for upload
shell: bash
run: |
mkdir -p staging/bin staging/lib
cp target/release_debug/data-gen staging/bin/
cp target/release_debug/datafusion-bench staging/bin/
cp target/release_debug/duckdb-bench staging/bin/
if [ -f target/release_debug/lance-bench ]; then
cp target/release_debug/lance-bench staging/bin/
fi
cp target/duckdb-lib-*/libduckdb.so staging/lib/ 2>/dev/null || true
- uses: actions/upload-artifact@v7
with:
name: sql-bench-binaries
path: staging/
retention-days: 1

bench:
needs: build
timeout-minutes: 120
strategy:
fail-fast: false
Expand All @@ -112,8 +164,6 @@ jobs:
steps:
- uses: runs-on/action@v2
if: inputs.mode != 'pr' || github.event.pull_request.head.repo.fork == false
with:
sccache: s3
- uses: actions/checkout@v6
if: inputs.mode == 'pr'
with:
Expand All @@ -123,9 +173,6 @@ jobs:
if: inputs.mode != 'pr'
- name: Setup benchmark environment
run: sudo bash scripts/setup-benchmark.sh
- uses: ./.github/actions/setup-rust
with:
repo-token: ${{ secrets.GITHUB_TOKEN }}

- name: Install DuckDB
run: |
Expand All @@ -135,16 +182,17 @@ jobs:

- uses: ./.github/actions/system-info

- name: Build binaries
- uses: actions/download-artifact@v8
with:
name: sql-bench-binaries
path: staging
- name: Install binaries
shell: bash
env:
RUSTFLAGS: "-C target-cpu=native -C force-frame-pointers=yes"
run: |
packages="--bin data-gen --bin datafusion-bench --bin duckdb-bench"
if [ "${{ matrix.build_lance }}" = "true" ]; then
packages="$packages --bin lance-bench"
fi
cargo build $packages --profile release_debug
mkdir -p target/release_debug
cp staging/bin/* target/release_debug/
chmod +x target/release_debug/*
echo "LD_LIBRARY_PATH=${{ github.workspace }}/staging/lib" >> "$GITHUB_ENV"

- name: Generate data
shell: bash
Expand Down
Loading