Zachery Aaron Shores-Chmielewski
b23f82e9a9
Structured observability for the iroh/SWIM layer: Aggregator, typed Event/Snapshot types, Sink (NoopSink default), ProbeScheduler, process stats, and host/iroh/swim introspection, plus the swactor-diag-collector, -postproc, and -iroh-relay binaries that assemble and render per-run bundles. Generalizes the pipeline-parallel-inference example to N stages and adds the topology-planner spec. Signed-off-by: Zachery Aaron Shores-Chmielewski <zacheryasc@gmail.com>
28 lines
1.1 KiB
Text
28 lines
1.1 KiB
Text
# CPU-only stub-mode image for the Docker-coordinated localised E2E
|
|
# (TEST_SPEC §13b). Runs the existing `pp-gpu-node` binary with the
|
|
# Python worker forced into stub mode (`PP_WORKER_STUB=1`) so the image
|
|
# needs neither CUDA nor tinygrad — just python3 to exec the worker.
|
|
#
|
|
# Build context must be the workspace root (the COPY paths assume that):
|
|
#
|
|
# docker build -f examples/pipeline-parallel-inference/Dockerfile.stub \
|
|
# -t pp-gpu-node-stub:latest .
|
|
#
|
|
# The host must have built the binaries first (release profile) — this
|
|
# image only packages them, it does not compile inside the container.
|
|
# `scripts/docker-e2e.sh` automates the full build + run.
|
|
FROM debian:trixie-slim
|
|
|
|
RUN apt-get update && \
|
|
apt-get install -y --no-install-recommends \
|
|
python3 \
|
|
ca-certificates && \
|
|
rm -rf /var/lib/apt/lists/*
|
|
|
|
COPY examples/pipeline-parallel-inference/target/release/pp-gpu-node /usr/local/bin/pp-gpu-node
|
|
COPY examples/pipeline-parallel-inference/pp_tinygrad_worker.py /usr/local/share/pp_tinygrad_worker.py
|
|
|
|
ENV WORKER_SCRIPT=/usr/local/share/pp_tinygrad_worker.py
|
|
ENV PP_WORKER_STUB=1
|
|
|
|
CMD ["pp-gpu-node"]
|