-
Notifications
You must be signed in to change notification settings - Fork 21.4k
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.
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
[PT-D][Checkpoint]Add MultiThreaded FileSystemWriter for distributed checkpointing and Update tests #87987
Conversation
🔗 Helpful Links🧪 See artifacts and rendered test results at hud.pytorch.org/pr/87987
Note: Links to docs will display an error until the docs builds have been completed. ✅ No FailuresAs of commit ac5583a: This comment was automatically generated by Dr. CI and updates every 15 minutes. |
Successfully rebased |
@pytorchmergebot merge |
Merge startedYour change will be merged once all checks pass (ETA 0-4 Hours). Learn more about merging in the wiki. Questions? Feedback? Please reach out to the PyTorch DevX Team |
…checkpointing and Update tests (pytorch#87987) This PR includes: Changes from @kumpera (pytorch#86327): adding MultiThreaded FileSystemWriter for distributed checkpointing, which adds two knobs to FileSystemWriter: thread_count and per_thread_copy_ahead. This increases up to 50% performance improvement on 32 GPUS workloads on AWS. Add parametrize tests to /test/distributed/_shard/checkpoint/test_file_system_checkpoint.py and /test/distributed/_shard/checkpoint/test_file_system_checkpoint_cpu.py Modify @with_comms in ShardedTensorTestBase to take in *args and **kwargs. Tests: ``` python3 test/distributed/checkpoint/test_file_system_checkpoint_cpu.py ``` test/distributed/checkpoint/test_file_system_checkpoint.py(GPU tests) runs fine locally but would timeout on CI. We will use thread-based PG and update this test in following PR. [T134844615] ## Add docstring and update comments in the following PRs. Pull Request resolved: pytorch#87987 Approved by: https://github.com/fduwjj
This PR includes:
Changes from @kumpera (#86327): adding MultiThreaded FileSystemWriter for distributed checkpointing, which adds two knobs to FileSystemWriter: thread_count and per_thread_copy_ahead. This increases up to 50% performance improvement on 32 GPUS workloads on AWS.
Add parametrize tests to /test/distributed/_shard/checkpoint/test_file_system_checkpoint.py and /test/distributed/_shard/checkpoint/test_file_system_checkpoint_cpu.py
Modify @with_comms in ShardedTensorTestBase to take in *args and **kwargs.
Tests:
test/distributed/checkpoint/test_file_system_checkpoint.py(GPU tests) runs fine locally but would timeout on CI. We will use thread-based PG and update this test in following PR.
[T134844615]
Add docstring and update comments in the following PRs.
cc @VitalyFedyunin @jgong5 @mingfeima @XiaobingSuper @sanchitintel @ashokei @jingxu10