Skip to content

Adds configurations for running p2p and Pathways-based checkpoint manager benchmarks on multi-slice v5litepod-16 TPUs. #3137

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

Open
copybara-service wants to merge 1 commit into
base: main
Choose a base branch
from
Open

Adds configurations for running p2p and Pathways-based checkpoint manager benchmarks on multi-slice v5litepod-16 TPUs. #3137

Changes from all commits
Commits
Show all changes
1 commit
Select commit
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
32 changes: 32 additions & 0 deletions checkpoint/orbax/checkpoint/_src/testing/oss/cloud_run_integration_tests.yaml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -15,3 +15,35 @@ tests:
verbose: true
skip_validation: true
enable_pathways: false
- name: p2p_checkpoint_manager_benchmark
cluster_name: orbax-cluster-test-mtc
tpu_type: v5litepod-16
zone: us-west1-c
gcp_region: us-west1
gcp_project: orbax-checkpoint
config_file: orbax/checkpoint/_src/testing/benchmarks/configs/p2p_checkpoint_manager_benchmark.yaml
docker_image: gcr.io/orbax-checkpoint/orbax-benchmarks-runner:latest
output_directory: gs://orbax-benchmarks/cloud_runs/
nodelete_cluster_on_completion: true
ramdisk_directory: /local/test
num_slices: 2
test_restart_workflow: false
verbose: true
skip_validation: true
enable_pathways: false
- name: pathways_checkpoint_manager_benchmark
cluster_name: orbax-cluster-test-pathways
tpu_type: v5litepod-16
zone: us-west1-c
gcp_region: us-west1
gcp_project: orbax-checkpoint
config_file: orbax/checkpoint/_src/testing/benchmarks/configs/checkpoint_manager_benchmark.yaml
docker_image: gcr.io/orbax-checkpoint/orbax-benchmarks-runner:latest
output_directory: gs://orbax-benchmarks/cloud_runs/
nodelete_cluster_on_completion: true
ramdisk_directory: /local/test
num_slices: 2
test_restart_workflow: false
verbose: true
skip_validation: true
enable_pathways: true
Loading