Rank local checkpointing in DCP internal without collectives#989
Rank local checkpointing in DCP internal without collectives#989saumishr wants to merge 1 commit intometa-pytorch:masterfrom
Conversation
|
This pull request was exported from Phabricator. Differential Revision: D72390326 |
…torch#989) Summary: ### Context DCP metadata collectives become prohibitively expensive as the job scale grows. This PR introduces rank-local checkpointing (XLFormers style checkpointing) which basically saves and loads the checkpoint without any collective. The trade off for now is the dedupe and re-sharding. Support for these would be introduced soon. Differential Revision: D72390326
8425538 to
21b56df
Compare
…torch#989) Summary: Pull Request resolved: meta-pytorch#989 ### Context DCP metadata collectives become prohibitively expensive as the job scale grows. This PR introduces rank-local checkpointing (XLFormers style checkpointing) which basically saves and loads the checkpoint without any collective. The trade off for now is the dedupe and re-sharding. Support for these would be introduced soon. Differential Revision: D72390326
|
This pull request was exported from Phabricator. Differential Revision: D72390326 |
21b56df to
94cc621
Compare
|
Hi @saumishr! Thank you for your pull request. We require contributors to sign our Contributor License Agreement, and yours needs attention. You currently have a record in our system, but the CLA is no longer valid, and will need to be resubmitted. ProcessIn order for us to review and merge your suggested changes, please sign at https://code.facebook.com/cla. If you are contributing on behalf of someone else (eg your employer), the individual CLA may not be sufficient and your employer may need to sign the corporate CLA. Once the CLA is signed, our tooling will perform checks and validations. Afterwards, the pull request will be tagged with If you have received this in error or have any questions, please contact us at cla@meta.com. Thanks! |
Summary:
Context
DCP metadata collectives become prohibitively expensive as the job scale grows. This PR introduces rank-local checkpointing (XLFormers style checkpointing) which basically saves and loads the checkpoint without any collective. The trade off for now is the dedupe and re-sharding. Support for these would be introduced soon.
Differential Revision: D72390326