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