Skip to content

Update zone for GPU test #1100

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
FIoannides merged 17 commits into from
Mar 23, 2026
Merged

Update zone for GPU test #1100

Changes from all commits
Commits
Show all changes
17 commits
Select commit Hold shift + click to select a range
a7ad78e
Update zone for GPU test
FIoannides Mar 6, 2026
9d9e025
update the machine type
FIoannides Mar 9, 2026
f8c0489
change region and device type
FIoannides Mar 11, 2026
083b11d
Update jobset used by GPUs from version 0.7.2 to 0.8.1
FIoannides Mar 13, 2026
13d0a37
revert back to b200 and use new us-east7-b zone
FIoannides Mar 16, 2026
2a8ed6b
change region and device type
FIoannides Mar 11, 2026
fe8788c
Update zone for GPU test
FIoannides Mar 6, 2026
9ad86e1
update the machine type
FIoannides Mar 9, 2026
7652c0c
change region and device type
FIoannides Mar 11, 2026
0bc0c73
Update jobset used by GPUs from version 0.7.2 to 0.8.1
FIoannides Mar 13, 2026
d549ae5
revert back to b200 and use new us-east7-b zone
FIoannides Mar 16, 2026
dba6fe3
change region and device type
FIoannides Mar 11, 2026
168e7cf
Merge branch 'fix-gpu-tests' of github.com:AI-Hypercomputer/xpk into …
FIoannides Mar 19, 2026
a74eafc
Merge branch 'main' into fix-gpu-tests
FIoannides Mar 20, 2026
4bbc261
Merge branch 'main' into fix-gpu-tests
FIoannides Mar 23, 2026
8252c22
Revert "Update jobset used by GPUs from version 0.7.2 to 0.8.1"
FIoannides Mar 23, 2026
dfad1e2
Merge branch 'main' into fix-gpu-tests
FIoannides Mar 23, 2026
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
16 changes: 8 additions & 8 deletions .github/workflows/integration_gpu_cluster_create.yaml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -27,7 +27,7 @@ jobs:
group: nightly-test-cluster-group-gpu
cancel-in-progress: false
env:
GPU_CLUSTER_NAME: nightly-xpk-b200
GPU_CLUSTER_NAME: nightly-xpk-h100
WORKLOAD_NAME: xpktest-gpu-nightly-${{ github.run_attempt }}
steps:
- uses: actions/download-artifact@v4
Expand Down Expand Up @@ -55,21 +55,21 @@ jobs:

# 4. Set Env Var for the host (GitHub Runner)
echo "GOOGLE_APPLICATION_CREDENTIALS=$HOME/.config/gcloud/application_default_credentials.json" >> $GITHUB_ENV
- name: Create an XPK Cluster with 1 x b200 GPU
run: xpk cluster create --cluster $GPU_CLUSTER_NAME --device-type=b200-8 --zone=asia-northeast1-b --default-pool-cpu-machine-type=n1-standard-16 --spot
- name: Create an XPK Cluster with 1 x h100 GPU
run: xpk cluster create --cluster $GPU_CLUSTER_NAME --device-type=h100-mega-80gb-8 --zone=asia-southeast1-b --default-pool-cpu-machine-type=e2-standard-8 --spot
- name: Authenticate Docker
run: gcloud auth configure-docker --quiet
- name: Run a base-docker-image workload
run: xpk workload create --cluster $GPU_CLUSTER_NAME --workload $WORKLOAD_NAME --docker-image='nvidia/cuda:12.1.0-base-ubuntu22.04' --command "nvidia-smi" --zone=asia-northeast1-b --device-type=b200-8
run: xpk workload create --cluster $GPU_CLUSTER_NAME --workload $WORKLOAD_NAME --docker-image='nvidia/cuda:12.1.0-base-ubuntu22.04' --command "nvidia-smi" --zone=asia-southeast1-b --device-type=h100-mega-80gb-8
- name: List out the workloads on the cluster
run: xpk workload list --cluster $GPU_CLUSTER_NAME --zone=asia-northeast1-b
run: xpk workload list --cluster $GPU_CLUSTER_NAME --zone=asia-southeast1-b
- name: Wait for workload completion and confirm it succeeded
run: xpk workload list --cluster $GPU_CLUSTER_NAME --zone=asia-northeast1-b --wait-for-job-completion $WORKLOAD_NAME --timeout 600
run: xpk workload list --cluster $GPU_CLUSTER_NAME --zone=asia-southeast1-b --wait-for-job-completion $WORKLOAD_NAME --timeout 600
- name: Delete the workload on the cluster
run: xpk workload delete --workload $WORKLOAD_NAME --cluster $GPU_CLUSTER_NAME --zone=asia-northeast1-b
run: xpk workload delete --workload $WORKLOAD_NAME --cluster $GPU_CLUSTER_NAME --zone=asia-southeast1-b
- name: Delete the cluster created
if: always()
run: xpk cluster delete --cluster $GPU_CLUSTER_NAME --zone=asia-northeast1-b --force
run: xpk cluster delete --cluster $GPU_CLUSTER_NAME --zone=asia-southeast1-b --force
- name: Upload cluster nodepool creation log
if: always()
uses: actions/upload-artifact@v4
Expand Down
Loading