Skip to content

Clean up Docker images on HPU and CPU benchmarks #106

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
huydhn merged 2 commits into from
Nov 14, 2025
Merged

Clean up Docker images on HPU and CPU benchmarks #106

Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
52bc890
[no ci] Clean up Docker images on HPU and CPU benchmarks
huydhn Nov 13, 2025
803fcaa
[no ci] Minor tweak
huydhn Nov 13, 2025
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
6 changes: 6 additions & 0 deletions .github/workflows/vllm-benchmark.yml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -191,6 +191,12 @@ jobs:
with:
registry-type: public

Copy link
Contributor

@yangw-dev yangw-dev Nov 14, 2025 •
edited
Loading

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

why it is download in the first place, if those docker images are unused at this point? It just download all of them , and only use some? curious about this, but will approve

Copy link
Contributor Author

@huydhn huydhn Nov 14, 2025

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

They are from previous jobs running on the same runner. This is a common issue for non-ephemeral runners where docker pull keeps piling up the images it uses over time. On our pet instances like H100, we have a daemon to clean up these images running on the server, so you don't see this step on CI. In this case, HPU servers are from Intel, so they are not set up in the same way

- name: Clean up unused Docker images
if: env.DEVICE_NAME == 'hpu' || contains(matrix.runner, 'gnr')
run: |
set -eux
docker image prune -a -f || true

- name: Check for last benchmark commit
working-directory: vllm-benchmarks
env:
Expand Down
Loading