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, 04:30 UTC
[CI] Adding CPU docker pipeline (#11261)
Signed-off-by: Yuan Zhou <yuan.zhou@intel.com> Co-authored-by: Kevin H. Luu <kevin@anyscale.com>
Yuan committed
Dec 20, 2024 at 03:46 UTC
a985f7af9f7b249974b283a9d999575ac30fac3d
1 file changed
+15
.buildkite/release-pipeline.yaml
+15
@@ -55,3 +55,18 @@ steps:
55
password-env: DOCKERHUB_TOKEN
56
env:
57
DOCKER_BUILDKIT: "1"
58
+
59
+ - block: "Build CPU release image"
60
+ key: block-cpu-release-image-build
61
+ depends_on: ~
62
+
63
+ - label: "Build and publish CPU release image"
64
+ depends_on: block-cpu-release-image-build
65
+ agents:
66
+ queue: cpu_queue_postmerge
67
+ commands:
68
+ - "aws ecr-public get-login-password --region us-east-1 | docker login --username AWS --password-stdin public.ecr.aws/q9t5s3a7"
69
+ - "DOCKER_BUILDKIT=1 docker build --build-arg max_jobs=16 --build-arg GIT_REPO_CHECK=1 --tag public.ecr.aws/q9t5s3a7/vllm-cpu-release-repo:$RELEASE_VERSION --progress plain -f Dockerfile.cpu ."
70
+ - "docker push public.ecr.aws/q9t5s3a7/vllm-cpu-release-repo:$RELEASE_VERSION"
71
+ env:
72
+ DOCKER_BUILDKIT: "1"