Skip to content

Commit 8120a78

Browse files
authored
Update TCPXO for A3 Mega GKE NCCL Test to run based on latest images (GoogleCloudPlatform#5154)
1 parent 7cd048c commit 8120a78

2 files changed

Lines changed: 3 additions & 3 deletions

File tree

modules/compute/gke-node-pool/gpu_direct.tf

Lines changed: 2 additions & 2 deletions
Original file line numberDiff line numberDiff line change
@@ -43,8 +43,8 @@ locals {
4343
"a3-megagpu-8g" = {
4444
# Manifest to be installed for enabling TCPXO on a3-megagpu-8g machines
4545
gpu_direct_manifests = [
46-
"https://raw.githubusercontent.com/GoogleCloudPlatform/container-engine-accelerators/bd4a7491672b48dfec28f3679b679a614f6cbbc7/gpudirect-tcpxo/nccl-tcpxo-installer.yaml", # nccl_plugin v1.0.14 for tcpxo
47-
"https://raw.githubusercontent.com/GoogleCloudPlatform/container-engine-accelerators/bd4a7491672b48dfec28f3679b679a614f6cbbc7/nri_device_injector/nri-device-injector.yaml", # nri_plugin
46+
"https://raw.githubusercontent.com/GoogleCloudPlatform/container-engine-accelerators/a32cd5152c6ac430ed0651199484853baa14f2d0/gpudirect-tcpxo/nccl-tcpxo-installer.yaml", # nccl_plugin v1.0.14 for tcpxo
47+
"https://raw.githubusercontent.com/GoogleCloudPlatform/container-engine-accelerators/a32cd5152c6ac430ed0651199484853baa14f2d0/nri_device_injector/nri-device-injector.yaml", # nri_plugin
4848
]
4949
updated_workload_path = replace(local.workload_path_tcpxo, ".yaml", "-tcpxo.yaml")
5050
rxdm_version = "v1.0.20" # matching nccl-tcpxo-installer version v1.0.14

tools/cloud-build/daily-tests/ansible_playbooks/test-validation/test-gke-a3-mega.yml

Lines changed: 1 addition & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -27,7 +27,7 @@
2727
delegate_to: localhost
2828
ansible.builtin.shell: |
2929
wget -O {{ workspace }}/examples/nccl-test.yaml \
30-
https://raw.githubusercontent.com/GoogleCloudPlatform/container-engine-accelerators/bd4a7491672b48dfec28f3679b679a614f6cbbc7/gpudirect-tcpxo/nccl-test-latest.yaml
30+
https://raw.githubusercontent.com/GoogleCloudPlatform/container-engine-accelerators/a32cd5152c6ac430ed0651199484853baa14f2d0/gpudirect-tcpxo/nccl-test-latest.yaml
3131
cat {{ workspace }}/examples/nccl-test.yaml
3232
register: nccl_test_file_contents
3333

0 commit comments

Comments
 (0)