build: Add support for sglang runtime image (#1770)

container/Dockerfile.sglang

@@ -74,7 +74,32 @@ RUN apt-get update -y && \
 
 WORKDIR /workspace
 
-### TODO: Bring back UCX EFA setup once we are confident it works with IB devices
+### UCX EFA Setup ###
+RUN rm -rf /opt/hpcx/ucx
+RUN rm -rf /usr/local/ucx
+RUN cd /usr/local/src && \
+    git clone https://github.com/openucx/ucx.git && \
+    cd ucx &&                   \
+    git checkout v1.19.x &&     \
+    ./autogen.sh && ./configure \
+    --prefix=/usr/local/ucx     \
+    --enable-shared             \
+    --disable-static            \
+    --disable-doxygen-doc       \
+    --enable-optimizations      \
+    --enable-cma                \
+    --enable-devel-headers      \
+    --with-cuda=/usr/local/cuda \
+    --with-verbs                \
+    --with-efa                  \
+    --with-dm                   \
+    --with-gdrcopy=/usr/local   \
+    --enable-mt &&              \
+    make -j &&                  \
+    make -j install-strip &&    \
+    ldconfig
+
+ENV LD_LIBRARY_PATH=/usr/lib:/usr/local/ucx/lib:$LD_LIBRARY_PATH
 ENV CPATH=/usr/include:$CPATH
 ENV PATH=/usr/bin:$PATH
 ENV PKG_CONFIG_PATH=/usr/lib/pkgconfig:$PKG_CONFIG_PATH
@@ -127,21 +152,22 @@ ENV VIRTUAL_ENV=/opt/dynamo/venv
 ENV PATH="${VIRTUAL_ENV}/bin:${PATH}"
 
 # Install NIXL Python module
-# TODO: Move gds_path selection based on arch into NIXL build
-RUN if [ "$ARCH" = "arm64" ]; then \
-        cd /opt/nixl && uv pip install . --config-settings=setup-args="-Dgds_path=/usr/local/cuda/targets/sbsa-linux"; \
-    else \
-        cd /opt/nixl && uv pip install . ; \
-    fi
+RUN cd /opt/nixl && uv build . --out-dir /workspace/wheels/nixl
+
+# Install the wheel
+# TODO: Move NIXL wheel install to the wheel_builder stage
+RUN uv pip install /workspace/wheels/nixl/*.whl
 
 # Install sglang
-# This commit references a NIXL fix that was releasted after the 0.4.8.post1 release https://github.com/sgl-project/sglang/pull/7330
+# This commit references a NIXL fix that was released after the 0.4.8.post1 release https://github.com/sgl-project/sglang/pull/7330
+#TODO: Built wheel should become an artifact which can be cached and reused in subsequent builds
 ARG SGLANG_COMMIT="bb9b608c86ebad7d9d01e29fe058bc184dc7285f"
 RUN --mount=type=cache,target=/root/.cache/uv \
     cd /opt && \
     git clone https://github.com/sgl-project/sglang.git && \
     cd sglang && \
     git checkout ${SGLANG_COMMIT} && \
+    # Install in editable mode for development
     uv pip install -e "python[all]"
 
 # Set env var that allows for forceful shutdown of inflight requests in SGL's TokenizerManager
@@ -380,20 +406,43 @@ ENV DYNAMO_HOME=/workspace
 ENV VIRTUAL_ENV=/opt/dynamo/venv
 ENV PATH="${VIRTUAL_ENV}/bin:${PATH}"
 
+### COPY NATS & ETCD ###
+# Copy nats and etcd from base image
+COPY --from=base /usr/bin/nats-server /usr/bin/nats-server
+COPY --from=base /usr/local/bin/etcd/ /usr/local/bin/etcd/
+ENV PATH=/usr/local/bin/etcd/:$PATH
+
+# Copy UCX from base image as plugin for NIXL
+# Copy NIXL source from base image (required for NIXL plugins)
+COPY --from=base /usr/local/ucx /usr/local/ucx
+COPY --from=base /usr/local/nixl /usr/local/nixl
+ARG ARCH_ALT
+ENV NIXL_PLUGIN_DIR=/usr/local/nixl/lib/${ARCH_ALT}-linux-gnu/plugins
+ENV LD_LIBRARY_PATH=/usr/local/nixl/lib/${ARCH_ALT}-linux-gnu:/usr/local/nixl/lib/${ARCH_ALT}-linux-gnu/plugins:/usr/local/ucx/lib:$LD_LIBRARY_PATH
+
 # Setup the python environment
 COPY --from=ghcr.io/astral-sh/uv:latest /uv /uvx /bin/
 RUN apt-get update && \
-    DEBIAN_FRONTEND=noninteractive apt-get install -y --no-install-recommends python3-dev && \
+    DEBIAN_FRONTEND=noninteractive apt-get install -y --no-install-recommends build-essential python3-dev libnuma-dev && \
     rm -rf /var/lib/apt/lists/* && \
     uv venv $VIRTUAL_ENV --python 3.12 && \
     echo "source $VIRTUAL_ENV/bin/activate" >> ~/.bashrc
 
+# Install SGLang and related packages (sgl-kernel, einops, sentencepiece) since they are not included in the runtime wheel
+# https://github.com/sgl-project/sglang/blob/v0.4.9.post1/python/pyproject.toml#L18-51
+RUN uv pip install "sglang[runtime_common]>=0.4.9.post1" && \
+    uv pip install einops && \
+    uv pip install sgl-kernel==0.2.4 && \
+    uv pip install sentencepiece
+
 # Install the wheels and symlink executables to /usr/local/bin so dynamo components can use them
 # Dynamo components currently do not have the VIRTUAL_ENV in their PATH, so we need to symlink the executables
 COPY --from=wheel_builder /workspace/dist/*.whl wheelhouse/
-RUN uv pip install ai-dynamo[vllm] --find-links wheelhouse && \
-    ln -sf $VIRTUAL_ENV/bin/* /usr/local/bin/ && \
-    rm -r wheelhouse
+COPY --from=base /workspace/wheels/nixl/*.whl wheelhouse/
+RUN uv pip install ai-dynamo --find-links wheelhouse && \
+    uv pip install ai-dynamo-runtime --find-links wheelhouse && \
+    uv pip install nixl --find-links wheelhouse && \
+    ln -sf $VIRTUAL_ENV/bin/* /usr/local/bin/
 
 # Tell vllm to use the Dynamo LLM C API for KV Cache Routing
 ENV VLLM_KV_CAPI_PATH="/opt/dynamo/bindings/lib/libdynamo_llm_capi.so"
@@ -403,8 +452,9 @@ RUN --mount=type=bind,source=./container/launch_message.txt,target=/workspace/la
     sed '/^#\s/d' /workspace/launch_message.txt > ~/.launch_screen && \
     echo "cat ~/.launch_screen" >> ~/.bashrc
 
-# Copy examples
-COPY ./examples examples/
+# Copy examples and set up Python path
+COPY . /workspace
+ENV PYTHONPATH=/workspace/examples/sglang/utils:$PYTHONPATH
 
-ENTRYPOINT [ "/usr/bin/bash" ]
+ENTRYPOINT ["/opt/nvidia/nvidia_entrypoint.sh"]
 CMD []