Tiny refactor router benchmark github actions (#16374)

.github/workflows/pr-benchmark-rust.yml

@@ -25,7 +25,6 @@ permissions:
   issues: write
 
 jobs:
-  # Quick check job that always runs on PRs
   benchmark-compile-check:
     name: Benchmark Compilation Check
     runs-on: ubuntu-latest
@@ -47,9 +46,7 @@ jobs:
         uses: Swatinem/rust-cache@v2
         with:
           workspaces: sgl-model-gateway
-          # Share cache across all benchmark jobs
           shared-key: "rust-cache"
-          # Save cache even on failure
           save-if: true
           cache-all-crates: true
           cache-on-failure: true
@@ -64,92 +61,47 @@ jobs:
         if: always()
         run: sccache --show-stats
 
-  # Full benchmark jobs that only run with label or on main branch
+  benchmark:
-  benchmark-request-processing:
+    name: Benchmark - ${{ matrix.name }}
-    name: Request Processing Benchmark
     if: |
       github.repository == 'sgl-project/sglang' &&
       (github.event_name == 'push' ||
        github.event_name == 'workflow_dispatch' ||
        (contains(github.event.pull_request.labels.*.name, 'router-benchmark') &&
         contains(github.event.pull_request.labels.*.name, 'run-ci')))
-    runs-on: ubuntu-latest
+    strategy:
-    steps:
+      fail-fast: false
-      - name: Checkout code
+      matrix:
-        uses: actions/checkout@v4
+        include:
-        with:
+          - name: Request Processing
-          # Fetch enough history for baseline comparison
+            bench_name: request_processing
-          fetch-depth: 100
+            bench_args: "benchmark_summary --exact"
-
+            runner: ubuntu-latest
-      - name: Install dependencies
+            sccache_version: "v0.12.0"
-        run: |
+            artifact_name: request-processing-results
-          bash scripts/ci/ci_install_rust.sh
+            artifact_path: criterion/benchmark_summary/
-
+          - name: Tokenizer
-      - name: Configure sccache
+            bench_name: tokenizer_benchmark
-        uses: mozilla-actions/sccache-action@v0.0.9
+            bench_args: ""
-        with:
+            runner: ubuntu-latest
-          version: "v0.12.0"
+            sccache_version: "v0.12.0"
-          disable_annotations: true
+            artifact_name: tokenizer-results
-
+            artifact_path: criterion/tokenizer*/
-      - name: Rust cache
+          - name: Tool Parser
-        uses: Swatinem/rust-cache@v2
+            bench_name: tool_parser_benchmark
-        with:
+            bench_args: ""
-          workspaces: sgl-model-gateway
+            runner: ubuntu-latest
-          # Share cache across all benchmark jobs
+            sccache_version: "v0.12.0"
-          shared-key: "rust-cache"
+            artifact_name: tool-parser-results
-          cache-all-crates: true
+            artifact_path: criterion/tool_parser*/
-          cache-on-failure: true
+          - name: Tree
-          # Save cache even on failure
+            bench_name: tree_benchmark
-          save-if: true
+            bench_args: "benchmark_summary"
-
+            runner: 4-gpu-a10
-      - name: Run request processing benchmark
+            sccache_version: "v0.10.0"
-        timeout-minutes: 30
+            artifact_name: tree-results
-        run: |
+            artifact_path: criterion/benchmark_summary*/
-          source "$HOME/.cargo/env"
+    runs-on: ${{ matrix.runner }}
-          cd sgl-model-gateway/
-          # Try to use sccache, but disable if it fails
-          if command -v sccache &> /dev/null; then
-            echo "Testing sccache availability..."
-            # Try to start sccache and check if it works
-            export RUSTC_WRAPPER=sccache
-            export SCCACHE_GHA_ENABLED="true"
-            if sccache --start-server 2>/dev/null && sccache --show-stats 2>/dev/null; then
-              echo "sccache is working, using it for compilation"
-            else
-              echo "sccache failed to start, falling back to regular cargo"
-              unset RUSTC_WRAPPER
-              unset SCCACHE_GHA_ENABLED
-            fi
-          else
-            echo "sccache not available, using regular cargo"
-          fi
-          # Run only the summary benchmark for quick validation in PRs
-          # Capture output to file for CI summary
-          cargo bench --bench request_processing -- benchmark_summary --exact 2>&1 | tee benchmark_output.txt
-
-      - name: Upload benchmark results
-        if: always()
-        uses: actions/upload-artifact@v4
-        with:
-          name: request-processing-results-${{ github.sha }}
-          path: |
-            sgl-model-gateway/target/criterion/benchmark_summary/
-            sgl-model-gateway/benchmark_output.txt
-          retention-days: 30
-
-      - name: Show sccache stats
-        if: always()
-        run: sccache --show-stats
-
-  benchmark-tokenizer:
-    name: Tokenizer Benchmark
-    if: |
-      github.repository == 'sgl-project/sglang' &&
-      (github.event_name == 'push' ||
-       github.event_name == 'workflow_dispatch' ||
-       (contains(github.event.pull_request.labels.*.name, 'router-benchmark') &&
-        contains(github.event.pull_request.labels.*.name, 'run-ci')))
-    runs-on: ubuntu-latest
     steps:
       - name: Checkout code
         uses: actions/checkout@v4
@@ -163,29 +115,25 @@ jobs:
       - name: Configure sccache
         uses: mozilla-actions/sccache-action@v0.0.9
         with:
-          version: "v0.12.0"
+          version: ${{ matrix.sccache_version }}
           disable_annotations: true
 
       - name: Rust cache
         uses: Swatinem/rust-cache@v2
         with:
           workspaces: sgl-model-gateway
-          # Share cache across all benchmark jobs
           shared-key: "rust-cache"
           cache-all-crates: true
           cache-on-failure: true
-          # Save cache even on failure
           save-if: true
 
-      - name: Run tokenizer benchmark
+      - name: Run benchmark
         timeout-minutes: 30
         run: |
           source "$HOME/.cargo/env"
           cd sgl-model-gateway/
-          # Try to use sccache, but disable if it fails
           if command -v sccache &> /dev/null; then
             echo "Testing sccache availability..."
-            # Try to start sccache and check if it works
             export RUSTC_WRAPPER=sccache
             export SCCACHE_GHA_ENABLED="true"
             if sccache --start-server 2>/dev/null && sccache --show-stats 2>/dev/null; then
@@ -198,163 +146,15 @@ jobs:
           else
             echo "sccache not available, using regular cargo"
           fi
-          # Capture output to file for CI summary
+          cargo bench --bench ${{ matrix.bench_name }} -- ${{ matrix.bench_args }} 2>&1 | tee benchmark_output.txt
-          cargo bench --bench tokenizer_benchmark 2>&1 | tee benchmark_output.txt
 
       - name: Upload benchmark results
         if: always()
         uses: actions/upload-artifact@v4
         with:
-          name: tokenizer-results-${{ github.sha }}
+          name: ${{ matrix.artifact_name }}-${{ github.sha }}
           path: |
-            sgl-model-gateway/target/criterion/tokenizer*/
+            sgl-model-gateway/target/${{ matrix.artifact_path }}
-            sgl-model-gateway/benchmark_output.txt
-          retention-days: 30
-
-  benchmark-tool-parser:
-    name: Tool Parser Benchmark
-    if: |
-      github.repository == 'sgl-project/sglang' &&
-      (github.event_name == 'push' ||
-       github.event_name == 'workflow_dispatch' ||
-       (contains(github.event.pull_request.labels.*.name, 'router-benchmark') &&
-        contains(github.event.pull_request.labels.*.name, 'run-ci')))
-    runs-on: ubuntu-latest
-    steps:
-      - name: Checkout code
-        uses: actions/checkout@v4
-        with:
-          fetch-depth: 100
-
-      - name: Install dependencies
-        run: |
-          bash scripts/ci/ci_install_rust.sh
-
-      - name: Configure sccache
-        uses: mozilla-actions/sccache-action@v0.0.9
-        with:
-          version: "v0.12.0"
-          disable_annotations: true
-
-      - name: Rust cache
-        uses: Swatinem/rust-cache@v2
-        with:
-          workspaces: sgl-model-gateway
-          # Share cache across all benchmark jobs
-          shared-key: "rust-cache"
-          cache-all-crates: true
-          cache-on-failure: true
-          # Save cache even on failure
-          save-if: true
-
-      - name: Run tool parser benchmark
-        timeout-minutes: 30
-        run: |
-          source "$HOME/.cargo/env"
-          cd sgl-model-gateway/
-          # Try to use sccache, but disable if it fails
-          if command -v sccache &> /dev/null; then
-            echo "Testing sccache availability..."
-            # Try to start sccache and check if it works
-            export RUSTC_WRAPPER=sccache
-            export SCCACHE_GHA_ENABLED="true"
-            if sccache --start-server 2>/dev/null && sccache --show-stats 2>/dev/null; then
-              echo "sccache is working, using it for compilation"
-            else
-              echo "sccache failed to start, falling back to regular cargo"
-              unset RUSTC_WRAPPER
-              unset SCCACHE_GHA_ENABLED
-            fi
-          else
-            echo "sccache not available, using regular cargo"
-          fi
-          # Capture output to file for CI summary
-          cargo bench --bench tool_parser_benchmark 2>&1 | tee benchmark_output.txt
-
-      - name: Upload benchmark results
-        if: always()
-        uses: actions/upload-artifact@v4
-        with:
-          name: tool-parser-results-${{ github.sha }}
-          path: |
-            sgl-model-gateway/target/criterion/tool_parser*/
-            sgl-model-gateway/benchmark_output.txt
-          retention-days: 30
-
-      - name: Show sccache stats
-        if: always()
-        run: sccache --show-stats
-
-  benchmark-tree:
-    name: Tree Benchmark
-    if: |
-      github.repository == 'sgl-project/sglang' &&
-      (github.event_name == 'push' ||
-       github.event_name == 'workflow_dispatch' ||
-       (contains(github.event.pull_request.labels.*.name, 'router-benchmark') &&
-        contains(github.event.pull_request.labels.*.name, 'run-ci')))
-    # Use 4-core runner for concurrent benchmarks that spawn multiple threads
-    runs-on: 4-gpu-a10
-    steps:
-      - name: Checkout code
-        uses: actions/checkout@v4
-        with:
-          fetch-depth: 100
-
-      - name: Install dependencies
-        run: |
-          bash scripts/ci/ci_install_rust.sh
-
-      - name: Configure sccache
-        uses: mozilla-actions/sccache-action@v0.0.9
-        with:
-          version: "v0.10.0"
-          disable_annotations: true
-
-      - name: Rust cache
-        uses: Swatinem/rust-cache@v2
-        with:
-          workspaces: sgl-model-gateway
-          # Share cache across all benchmark jobs
-          shared-key: "rust-cache"
-          cache-all-crates: true
-          cache-on-failure: true
-          # Save cache even on failure
-          save-if: true
-
-      - name: Run tree benchmark
-        timeout-minutes: 30
-        run: |
-          source "$HOME/.cargo/env"
-          cd sgl-model-gateway/
-          # Try to use sccache, but disable if it fails
-          if command -v sccache &> /dev/null; then
-            echo "Testing sccache availability..."
-            # Try to start sccache and check if it works
-            export RUSTC_WRAPPER=sccache
-            export SCCACHE_GHA_ENABLED="true"
-            if sccache --start-server 2>/dev/null && sccache --show-stats 2>/dev/null; then
-              echo "sccache is working, using it for compilation"
-            else
-              echo "sccache failed to start, falling back to regular cargo"
-              unset RUSTC_WRAPPER
-              unset SCCACHE_GHA_ENABLED
-            fi
-          else
-            echo "sccache not available, using regular cargo"
-          fi
-          # Run summary benchmark for quick validation in CI
-          # Note: Don't use --exact because benchmark names are prefixed (benchmark_summary/insert_realistic)
-          # Capture output to file for CI summary
-          cargo bench --bench tree_benchmark -- benchmark_summary 2>&1 | tee benchmark_output.txt
-
-      - name: Upload benchmark results
-        if: always()
-        uses: actions/upload-artifact@v4
-        with:
-          name: tree-results-${{ github.sha }}
-          path: |
-            sgl-model-gateway/target/criterion/benchmark_summary*/
             sgl-model-gateway/benchmark_output.txt
           retention-days: 30
 
@@ -364,7 +164,7 @@ jobs:
 
   benchmark-summary:
     name: Benchmark Summary
-    needs: [benchmark-request-processing, benchmark-tokenizer, benchmark-tool-parser, benchmark-tree]
+    needs: [benchmark]
     if: always() && (github.repository == 'sgl-project/sglang' || github.event_name == 'pull_request')
     runs-on: ubuntu-latest
     steps:
@@ -379,7 +179,6 @@ jobs:
           echo "## 🚀 Benchmark Results Summary" > summary.md
           echo "" >> summary.md
 
-          # Request Processing Benchmark
           echo "### Request Processing" >> summary.md
           REQ_DIR="benchmark-results/request-processing-results-${{ github.sha }}"
           if [ -d "$REQ_DIR" ]; then
@@ -390,7 +189,6 @@ jobs:
               echo "<summary>View Results</summary>" >> summary.md
               echo "" >> summary.md
               echo '```' >> summary.md
-              # Extract the summary section (starts with "SGLang Model Gateway" or performance data)
               grep -A 100 "SGLang Model Gateway\|Quick Performance\|Performance Insights" "$REQ_DIR/benchmark_output.txt" | head -50 >> summary.md || cat "$REQ_DIR/benchmark_output.txt" | tail -60 >> summary.md
               echo '```' >> summary.md
               echo "</details>" >> summary.md
@@ -400,7 +198,6 @@ jobs:
           fi
           echo "" >> summary.md
 
-          # Tokenizer Benchmark
           echo "### Tokenizer" >> summary.md
           TOK_DIR="benchmark-results/tokenizer-results-${{ github.sha }}"
           if [ -d "$TOK_DIR" ]; then
@@ -411,7 +208,6 @@ jobs:
               echo "<summary>View Results</summary>" >> summary.md
               echo "" >> summary.md
               echo '```' >> summary.md
-              # Extract the summary table section
               grep -A 200 "TOKENIZER BENCHMARK SUMMARY\|ENCODING THROUGHPUT" "$TOK_DIR/benchmark_output.txt" | head -100 >> summary.md || cat "$TOK_DIR/benchmark_output.txt" | tail -100 >> summary.md
               echo '```' >> summary.md
               echo "</details>" >> summary.md
@@ -421,7 +217,6 @@ jobs:
           fi
           echo "" >> summary.md
 
-          # Tool Parser Benchmark
           echo "### Tool Parser" >> summary.md
           TOOL_DIR="benchmark-results/tool-parser-results-${{ github.sha }}"
           if [ -d "$TOOL_DIR" ]; then
@@ -432,7 +227,6 @@ jobs:
               echo "<summary>View Results</summary>" >> summary.md
               echo "" >> summary.md
               echo '```' >> summary.md
-              # Extract the summary table section
               grep -A 200 "TOOL PARSER BENCHMARK SUMMARY\|REGISTRY OPERATIONS\|COMPLETE PARSING" "$TOOL_DIR/benchmark_output.txt" | head -100 >> summary.md || cat "$TOOL_DIR/benchmark_output.txt" | tail -100 >> summary.md
               echo '```' >> summary.md
               echo "</details>" >> summary.md
@@ -442,7 +236,6 @@ jobs:
           fi
           echo "" >> summary.md
 
-          # Tree Benchmark
           echo "### Radix Tree (Cache-Aware Routing)" >> summary.md
           TREE_DIR="benchmark-results/tree-results-${{ github.sha }}"
           if [ -d "$TREE_DIR" ]; then
@@ -453,7 +246,6 @@ jobs:
               echo "<summary>View Results</summary>" >> summary.md
               echo "" >> summary.md
               echo '```' >> summary.md
-              # Extract the full summary table (avoids duplication with [BENCH_RESULT] lines)
               grep -A 50 "RADIX TREE BENCHMARK SUMMARY" "$TREE_DIR/benchmark_output.txt" | head -50 >> summary.md || cat "$TREE_DIR/benchmark_output.txt" | tail -50 >> summary.md
               echo '```' >> summary.md
               echo "</details>" >> summary.md
@@ -466,10 +258,7 @@ jobs:
           echo "---" >> summary.md
           echo "_Generated at $(date -u '+%Y-%m-%d %H:%M:%S UTC')_" >> summary.md
 
-          # Display summary in logs
           cat summary.md
-
-          # Write to GitHub Step Summary for visibility in the Actions UI
           cat summary.md >> $GITHUB_STEP_SUMMARY
 
       - name: Upload summary