Build & publish the simulator image via GitHub Actions

.github/workflows/build-sim-image.yml

@@ -0,0 +1,107 @@
+name: Build simulator image
+
+# Builds an Ubuntu-based image with the simulator side fully installed
+# (ASTRA-Sim analytical backend, chakra, python deps — exactly what
+# scripts/install-sim.sh produces on a bare host) and pushes it to
+# GitHub Container Registry. PRs run the build to validate the install
+# path; only pushes to main (and manual dispatch) publish the image.
+
+on:
+  push:
+    branches: [main]
+    paths:
+      - 'scripts/install-sim.sh'
+      - 'scripts/compile.sh'
+      - 'scripts/sim.Dockerfile'
+      - 'astra-sim'
+      - '.gitmodules'
+      - '.github/workflows/build-sim-image.yml'
+  pull_request:
+    paths:
+      - 'scripts/install-sim.sh'
+      - 'scripts/compile.sh'
+      - 'scripts/sim.Dockerfile'
+      - 'astra-sim'
+      - '.gitmodules'
+      - '.github/workflows/build-sim-image.yml'
+  workflow_dispatch:
+
+permissions:
+  contents: read
+  packages: write
+
+env:
+  REGISTRY: ghcr.io
+
+# One in-flight build per ref; queued PR builds get cancelled in favour
+# of the newest commit. Main pushes finish.
+concurrency:
+  group: build-sim-image-${{ github.ref }}
+  cancel-in-progress: ${{ github.event_name == 'pull_request' }}
+
+jobs:
+  build:
+    runs-on: ubuntu-latest
+    steps:
+      - name: Checkout (with submodules)
+        uses: actions/checkout@v4
+        with:
+          submodules: recursive
+
+      # GHCR rejects uppercase characters in image names, but
+      # ${{ github.repository }} preserves the original case
+      # (PSAL-POSTECH/LLMServingSimSpec). Lowercase it once here so
+      # every downstream step sees a clean reference.
+      - name: Resolve image name (lowercase)
+        id: img
+        run: |
+          name="${GITHUB_REPOSITORY,,}/sim"
+          echo "ref=${REGISTRY}/${name}" >> "$GITHUB_OUTPUT"
+
+      - name: Set up Docker Buildx
+        uses: docker/setup-buildx-action@v3
+
+      - name: Log in to GHCR
+        if: github.event_name != 'pull_request'
+        uses: docker/login-action@v3
+        with:
+          registry: ${{ env.REGISTRY }}
+          username: ${{ github.actor }}
+          password: ${{ secrets.GITHUB_TOKEN }}
+
+      # latest → main only; sha-<short> → every build; branch tag for
+      # non-main branches so feature-branch images are addressable.
+      - name: Image tags + labels
+        id: meta
+        uses: docker/metadata-action@v5
+        with:
+          images: ${{ steps.img.outputs.ref }}
+          tags: |
+            type=raw,value=latest,enable={{is_default_branch}}
+            type=sha,prefix=sha-,format=short
+            type=ref,event=branch
+
+      - name: Build (and push if not PR)
+        uses: docker/build-push-action@v5
+        with:
+          context: .
+          file: scripts/sim.Dockerfile
+          push: ${{ github.event_name != 'pull_request' }}
+          tags: ${{ steps.meta.outputs.tags }}
+          labels: ${{ steps.meta.outputs.labels }}
+          cache-from: type=gha
+          cache-to: type=gha,mode=max
+
+      - name: Summary
+        if: always()
+        run: |
+          {
+            echo "### Simulator image"
+            echo
+            echo "- **image:** \`${{ steps.img.outputs.ref }}\`"
+            echo "- **tags:**"
+            echo '```'
+            echo "${{ steps.meta.outputs.tags }}"
+            echo '```'
+            echo "- **pushed:** ${{ github.event_name != 'pull_request' }}"
+          } >> "$GITHUB_STEP_SUMMARY"

scripts/README.md

@@ -14,6 +14,7 @@ live with their module — only setup and build helpers are here.
 | `install-vllm-cpu.sh` | Bare-metal vLLM **CPU** install. Builds vLLM 0.19.0 from source with `VLLM_TARGET_DEVICE=cpu` into `.venv-cpu` (separate from the GPU venv). Works on x86_64 and aarch64 (NVIDIA Grace). Used by `profiler/profile-cpu.sh`. |
 | `install-sim.sh`  | Bare-metal simulator install for fresh Docker containers / minimal Linux hosts (root or non-root). Installs apt build deps, inits submodules, pip-installs sim Python deps, builds ASTRA-Sim's analytical backend, and installs the Chakra converter (`--no-deps`). Skips vLLM — profiler/bench need `install-vllm.sh` and a GPU. Idempotent. |
 | `compile.sh`      | Build ASTRA-Sim's analytical backend and install the Chakra trace converter. |
+| `sim.Dockerfile`  | Ubuntu 24.04 base + `install-sim.sh`. Built and published to GHCR by `.github/workflows/build-sim-image.yml` on every push to `main` touching the install / build paths. Pull with `docker pull ghcr.io/psal-postech/llmservingsimspec/sim:latest`. |
 
 ## Typical first-time setup
 

scripts/sim.Dockerfile

@@ -0,0 +1,45 @@
+# LLMServingSim simulator image.
+#
+# Built by .github/workflows/build-sim-image.yml on every push to main
+# that touches scripts/install-sim.sh, the astra-sim submodule, or this
+# file. Published to ghcr.io/<owner>/<repo>/sim.
+#
+# Local build:
+#   docker build -f scripts/sim.Dockerfile -t llmservingsim-sim .
+#
+# Local run (mount your working copy over /workspace if you want live
+# edits — the image carries the repo state at build time):
+#   docker run -it --rm -v "$(pwd)":/workspace llmservingsim-sim bash
+
+FROM ubuntu:24.04
+
+ENV DEBIAN_FRONTEND=noninteractive \
+    PIP_BREAK_SYSTEM_PACKAGES=1
+
+# Pre-install the apt packages install-sim.sh expects. Doing this in a
+# dedicated layer lets Docker cache them independently of the script's
+# pip / submodule / build steps further down, which is what dominates
+# rebuild time when the source changes.
+RUN apt-get update -qq \
+ && apt-get install -y -qq --no-install-recommends \
+        git build-essential cmake \
+        protobuf-compiler libprotobuf-dev \
+        python3 python3-pip python-is-python3 \
+        ca-certificates curl \
+ && rm -rf /var/lib/apt/lists/*
+
+WORKDIR /workspace
+
+# Copy the repo (including submodules — the workflow's checkout step
+# fetches them on the runner so they're in the build context).
+COPY . /workspace
+
+# install-sim.sh is idempotent and detects pre-installed apt packages,
+# so its apt phase is a no-op here; the work it actually does in this
+# layer is: python deps, ASTRA-Sim analytical-backend build, chakra
+# install. See the script for the full sequence.
+RUN bash scripts/install-sim.sh
+
+# Default entry: interactive shell at the repo root. Match docker-sim.sh
+# behaviour so the two surfaces (CI image / dev script) feel the same.
+CMD ["/bin/bash"]