@tduncan / Nvidia-Optimization-Stack-VLLM Mirror · read-only / commits / 016e3676e7
Fork 0

A high-throughput and memory-efficient inference and serving engine for LLMs

Mirrored from https://github.com/tduncan-lum/Nvidia-Optimization-Stack---VLLM.git

Synced Nov 16, 02:00 UTC

[CI] add docker volume prune to neuron CI (#12291)

Signed-off-by: Liangfu Chen <liangfc@amazon.com>

Liangfu Chen committed Jan 21, 2025 at 18:47 UTC 016e3676e7eab58b177f92e579ba20173defee01
1 file changed +4 -1
.buildkite/run-neuron-test.sh
+4 -1
@@ -25,8 +25,11 @@ if [ -f /tmp/neuron-docker-build-timestamp ]; then
25 last_build=$(cat /tmp/neuron-docker-build-timestamp)
26 current_time=$(date +%s)
27 if [ $((current_time - last_build)) -gt 86400 ]; then
28 + # Remove dangling images (those that are not tagged and not used by any container)
29 docker image prune -f
29 - docker system prune -f
30 + # Remove unused volumes / force the system prune for old images as well.
31 + docker volume prune -f && docker system prune -f
32 + # Remove huggingface model artifacts and compiler cache
33 rm -rf "${HF_MOUNT:?}/*"
34 rm -rf "${NEURON_COMPILE_CACHE_MOUNT:?}/*"
35 echo "$current_time" > /tmp/neuron-docker-build-timestamp