Dockerfile.gpu

# GPU-enabled Compass build (cuVS / CAGRA → HNSW).
#
# Requires a Linux x86_64 host with CUDA 12+ runtime at deploy time.
# The first build pulls and compiles cuVS from source; expect 30-60 minutes
# on a fresh box. Cache the builder layer aggressively in CI.

# ── Stage 1: Build ────────────────────────────────────────────────────────────
FROM nvidia/cuda:12.4.1-devel-ubuntu22.04 AS builder

ENV DEBIAN_FRONTEND=noninteractive

RUN apt-get update && apt-get install -y \
    curl \
    git \
    cmake \
    pkg-config \
    libssl-dev \
    build-essential \
    gcc-11 \
    g++-11 \
    && rm -rf /var/lib/apt/lists/*

# Pin gcc-11 (cuVS requires gcc 11+).
RUN update-alternatives --install /usr/bin/gcc gcc /usr/bin/gcc-11 100 \
    && update-alternatives --install /usr/bin/g++ g++ /usr/bin/g++-11 100

# Install Rust 1.82.
RUN curl --proto '=https' --tlsv1.2 -sSf https://sh.rustup.rs | sh -s -- -y --default-toolchain 1.82.0
ENV PATH="/root/.cargo/bin:${PATH}"

WORKDIR /app

COPY Cargo.toml Cargo.lock* ./
COPY crates/ crates/

# Build with the gpu feature. cuVS will be cloned and cmake-built on first run.
RUN cargo build --release -p compass --features gpu

# ── Stage 2: Runtime ──────────────────────────────────────────────────────────
FROM nvidia/cuda:12.4.1-runtime-ubuntu22.04

RUN apt-get update && apt-get install -y \
    ca-certificates \
    && rm -rf /var/lib/apt/lists/*

WORKDIR /app

COPY --from=builder /app/target/release/compass /app/compass
RUN mkdir -p /app/data

ENV PORT=4001
ENV DATA_DIR=/app/data
ENV COMPASS_BACKEND=auto
EXPOSE 4001

CMD ["/app/compass"]