Abstract build step into composite action

.github/actions/build-commit/action.yaml

@@ -0,0 +1,81 @@
+name: Build a specific daft commit and store it in AWS S3
+description: Build a specific daft commit and store the output wheel in AWS S3
+
+inputs:
+  commit:
+    description: The commit hash to build
+    required: true
+
+outputs:
+  wheel:
+    description: The wheel file that was built
+    value: ${{ steps.upload_to_s3.outputs.wheel }}
+
+runs:
+  using: composite
+  steps:
+  - name: Assume AWS credentials
+    uses: aws-actions/configure-aws-credentials@v4
+    with:
+      aws-region: us-west-2
+      role-to-assume: ${{ secrets.ACTIONS_AWS_ROLE_ARN }}
+      role-session-name: daft-performance-comparisons-build
+  - name: Checkout commit ${{ inputs.commit }}
+    uses: actions/checkout@v4
+    with:
+      ref: ${{ inputs.commit }}
+  - uses: ./.github/actions/install
+  - name: Cache dependencies
+    uses: actions/cache@v3
+    with:
+      path: ~/target/debug
+      key: ${{ runner.os }}-cargo-deps-${{ hashFiles('**/Cargo.lock') }}
+      restore-keys: |
+        ${{ runner.os }}-cargo-deps-
+  - name: Build wheel for commit ${{ inputs.commit }}
+    shell: bash
+    run: |
+      export CARGO_TARGET_DIR=~/target
+      uv v
+      source .venv/bin/activate
+      uv pip install pip
+      uv pip install maturin
+      maturin build
+  - name: Upload files to s3
+    shell: bash
+    id: upload_to_s3
+    run: |
+      # There should only be one output wheel in this directory
+      for file in ~/target/wheels/*.whl; do
+        aws s3 cp $file s3://github-actions-artifacts-bucket/builds/${{ github.sha }}/${{ inputs.commit }}/ --acl public-read --no-progress;
+        file_basename=$(basename $file)
+        echo "wheel=$file_basename" >> "$GITHUB_OUTPUT"
+      done
+
+  # steps:
+  # - name: Install rust
+  #   shell: bash
+  #   run: |
+  #     curl --proto '=https' --tlsv1.2 -sSf https://sh.rustup.rs | sh -s -- -y
+  #     CARGO_BIN="$HOME/.cargo/bin"
+  #     echo 'export PATH="$CARGO_BIN:$PATH"' >> $HOME/.bashrc
+  #     echo "$CARGO_BIN" >> $GITHUB_PATH
+
+  # - name: Install uv
+  #   shell: bash
+  #   run: |
+  #     curl -LsSf https://astral.sh/uv/install.sh | sh
+  #     UV_BIN="$HOME/.local/bin"
+  #     echo 'export PATH="$UV_BIN:$PATH"' >> $HOME/.bashrc
+  #     echo "$UV_BIN" >> $GITHUB_PATH
+
+  # - name: Source .bashrc
+  #   shell: bash
+  #   run: |
+  #     source $HOME/.bashrc
+
+  # - name: Install python (version 3.9)
+  #   shell: bash
+  #   run: |
+  #     uv python install 3.9
+  #     uv python pin 3.9

.github/assets/ray.yaml

@@ -17,8 +17,7 @@ available_node_types:
     resources: {"CPU": 0}
     node_config:
       KeyName: ci-github-actions-ray-cluster-key
-      InstanceType: i3.2xlarge
-      ImageId: ami-0b8c6b923777519db
+      InstanceType: m6id.2xlarge
       IamInstanceProfile:
         Name: ray-autoscaler-v1
 
@@ -28,13 +27,21 @@ available_node_types:
     resources: {}
     node_config:
       KeyName: ci-github-actions-ray-cluster-key
-      InstanceType: i3.2xlarge
-      ImageId: ami-0b8c6b923777519db
+      InstanceType: m6id.2xlarge
       IamInstanceProfile:
         Name: ray-autoscaler-v1
 
 setup_commands:
+# Mount drive
+- sudo mkfs.ext4 /dev/nvme1n1
+- sudo mount -t ext4 /dev/nvme1n1 /tmp
+- sudo chmod 777 /tmp
+
 # Install dependencies
+- sudo apt install unzip
+- curl "https://awscli.amazonaws.com/awscli-exe-linux-x86_64.zip" -o "awscliv2.zip"
+- unzip awscliv2.zip
+- sudo ./aws/install
 - curl -LsSf https://astral.sh/uv/install.sh | sh
 - echo 'export PATH="$HOME/.local/bin:$PATH"' >> ~/.bashrc
 - source ~/.bashrc
@@ -44,6 +51,13 @@ setup_commands:
 - echo "source $HOME/.venv/bin/activate" >> $HOME/.bashrc
 - source .venv/bin/activate
 - uv pip install pip ray[default] py-spy
-    # GitHub Actions workflow will replace the `<<SHA>>`, `<<COMMIT>>`, and `<<WHEEL>>` placeholders
-    # with the actual values as determined dynamically during runtime of the actual workflow.
-- uv pip install https://github-actions-artifacts-bucket.s3.amazonaws.com/builds/<<SHA>>/<<COMMIT>>/<<WHEEL>>
+# GitHub Actions workflow will replace the `<<SHA>>`, `<<COMMIT>>`, and `<<WHEEL>>` placeholders
+# with the actual values as determined dynamically during runtime of the actual workflow.
+# - uv pip install https://github-actions-artifacts-bucket.s3.amazonaws.com/builds/<<SHA>>/<<COMMIT>>/<<WHEEL>>
+- uv pip install getdaft
+
+# Download benchmarking fixtures
+- |
+  aws s3 sync \
+  s3://eventual-dev-benchmarking-fixtures/uncompressed/tpch-dbgen/2_0/2/parquet/ \
+  /tmp/data/2_0/2/parquet/

.github/workflows/benchmark-local.yaml

@@ -26,45 +26,49 @@ jobs:
     outputs:
       wheel: ${{ steps.upload_to_s3.outputs.wheel }}
     steps:
-    - name: Assume AWS credentials
-      uses: aws-actions/configure-aws-credentials@v4
-      with:
-        aws-region: us-west-2
-        role-to-assume: ${{ secrets.ACTIONS_AWS_ROLE_ARN }}
-        role-session-name: daft-performance-comparisons-build
-    - name: Checkout commit ${{ matrix.commit }}
-      uses: actions/checkout@v4
+    - uses: ./.github/actions/build-commit
       with:
-        ref: ${{ matrix.commit }}
-    - uses: ./.github/actions/install
-    - name: Install additional dependencies
-      run: |
-        sudo apt-get update
-        sudo apt-get install gcc g++ pkg-config libssl-dev make -y
-    - name: Cache dependencies
-      uses: actions/cache@v3
-      with:
-        path: ~/target/debug
-        key: ${{ runner.os }}-cargo-deps-${{ hashFiles('**/Cargo.lock') }}
-        restore-keys: |
-          ${{ runner.os }}-cargo-deps-
-    - name: Build wheel for commit ${{ matrix.commit }}
-      run: |
-        export CARGO_TARGET_DIR=~/target
-        uv v
-        source .venv/bin/activate
-        uv pip install pip
-        uv pip install maturin
-        maturin build
-    - name: Upload files to s3
-      id: upload_to_s3
-      run: |
-        # There should only be one output wheel in this directory
-        for file in ~/target/wheels/*.whl; do
-          aws s3 cp $file s3://github-actions-artifacts-bucket/builds/${{ github.sha }}/${{ matrix.commit }}/ --acl public-read --no-progress;
-          file_basename=$(basename $file)
-          echo "wheel=$file_basename" >> "$GITHUB_OUTPUT"
-        done
+        commit: ${{ matrix.commit }}
+
+      # - name: Assume AWS credentials
+      #   uses: aws-actions/configure-aws-credentials@v4
+      #   with:
+      #     aws-region: us-west-2
+      #     role-to-assume: ${{ secrets.ACTIONS_AWS_ROLE_ARN }}
+      #     role-session-name: daft-performance-comparisons-build
+      # - name: Checkout commit ${{ matrix.commit }}
+      #   uses: actions/checkout@v4
+      #   with:
+      #     ref: ${{ matrix.commit }}
+      # - uses: ./.github/actions/install
+      # - name: Install additional dependencies
+      #   run: |
+      #     sudo apt-get update
+      #     sudo apt-get install gcc g++ pkg-config libssl-dev make -y
+      # - name: Cache dependencies
+      #   uses: actions/cache@v3
+      #   with:
+      #     path: ~/target/debug
+      #     key: ${{ runner.os }}-cargo-deps-${{ hashFiles('**/Cargo.lock') }}
+      #     restore-keys: |
+      #       ${{ runner.os }}-cargo-deps-
+      # - name: Build wheel for commit ${{ matrix.commit }}
+      #   run: |
+      #     export CARGO_TARGET_DIR=~/target
+      #     uv v
+      #     source .venv/bin/activate
+      #     uv pip install pip
+      #     uv pip install maturin
+      #     maturin build
+      # - name: Upload files to s3
+      #   id: upload_to_s3
+      #   run: |
+      #     # There should only be one output wheel in this directory
+      #     for file in ~/target/wheels/*.whl; do
+      #       aws s3 cp $file s3://github-actions-artifacts-bucket/builds/${{ github.sha }}/${{ matrix.commit }}/ --acl public-read --no-progress;
+      #       file_basename=$(basename $file)
+      #       echo "wheel=$file_basename" >> "$GITHUB_OUTPUT"
+      #     done
 
   run:
     needs: build