Rank local checkpointing in DCP without collectives #991
+6
−2
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
|
This pull request was exported from Phabricator. Differential Revision: D70112642 |
saumishr
added a commit
to saumishr/tnt
that referenced
this pull request
Apr 18, 2025
Summary: X-link: pytorch/pytorch#147758 Context: DCP metadata collectives become prohibitively expensive as the job scale grows. This PR introduces rank-local 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: D70112642
saumishr
force-pushed
the
export-D70112642
branch
from
058b3d6 to
d5af27e
Compare
|
This pull request was exported from Phabricator. Differential Revision: D70112642 |
1 similar comment
|
This pull request was exported from Phabricator. Differential Revision: D70112642 |
saumishr
added a commit
to saumishr/tnt
that referenced
this pull request
Apr 20, 2025
Summary: Pull Request resolved: pytorch#991 X-link: pytorch/pytorch#147758 Context: DCP metadata collectives become prohibitively expensive as the job scale grows. This PR introduces rank-local 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: D70112642
saumishr
force-pushed
the
export-D70112642
branch
from
d5af27e to
020eb87
Compare
pytorch-bot bot
pushed a commit
to pytorch/pytorch
that referenced
this pull request
Apr 24, 2025
Summary: X-link: pytorch/tnt#991 Context: DCP metadata collectives become prohibitively expensive as the job scale grows. This PR introduces rank-local 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. Test Plan: E2E UTs Differential Revision: D70112642
saumishr
force-pushed
the
export-D70112642
branch
from
020eb87 to
006e037
Compare
saumishr
added a commit
to saumishr/tnt
that referenced
this pull request
Apr 24, 2025
Summary: X-link: pytorch/pytorch#147758 Context: DCP metadata collectives become prohibitively expensive as the job scale grows. This PR introduces rank-local 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: D70112642
|
This pull request was exported from Phabricator. Differential Revision: D70112642 |
pytorch-bot bot
pushed a commit
to pytorch/pytorch
that referenced
this pull request
Apr 30, 2025
Summary: X-link: pytorch/tnt#991 Context: DCP metadata collectives become prohibitively expensive as the job scale grows. This PR introduces rank-local 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. Test Plan: E2E UTs Differential Revision: D70112642
saumishr
force-pushed
the
export-D70112642
branch
from
006e037 to
f8ad352
Compare
saumishr
added a commit
to saumishr/tnt
that referenced
this pull request
Apr 30, 2025
Summary: X-link: pytorch/pytorch#147758 Context: DCP metadata collectives become prohibitively expensive as the job scale grows. This PR introduces rank-local 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: D70112642
|
This pull request was exported from Phabricator. Differential Revision: D70112642 |
saumishr
force-pushed
the
export-D70112642
branch
from
f8ad352 to
76c5fcf
Compare
saumishr
added a commit
to saumishr/tnt
that referenced
this pull request
Jun 26, 2025
Summary: X-link: pytorch/pytorch#147758 Context: DCP metadata collectives become prohibitively expensive as the job scale grows. This PR introduces rank-local 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: D70112642
|
This pull request was exported from Phabricator. Differential Revision: D70112642 |
pytorch-bot bot
pushed a commit
to pytorch/pytorch
that referenced
this pull request
Jun 26, 2025
Summary: X-link: pytorch/tnt#991 Context: DCP metadata collectives become prohibitively expensive as the job scale grows. This PR introduces rank-local 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. Test Plan: E2E UTs Differential Revision: D70112642
saumishr
added a commit
to saumishr/pytorch
that referenced
this pull request
Jun 26, 2025
…ch#147758) Summary: X-link: pytorch/tnt#991 Context: DCP metadata collectives become prohibitively expensive as the job scale grows. This PR introduces rank-local 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. Test Plan: E2E UTs Differential Revision: D70112642
saumishr
added a commit
to saumishr/tnt
that referenced
this pull request
Jun 26, 2025
Summary: X-link: pytorch/pytorch#147758 Context: DCP metadata collectives become prohibitively expensive as the job scale grows. This PR introduces rank-local 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: D70112642
saumishr
force-pushed
the
export-D70112642
branch
from
76c5fcf to
b631684
Compare
|
This pull request was exported from Phabricator. Differential Revision: D70112642 |
pytorch-bot bot
pushed a commit
to pytorch/pytorch
that referenced
this pull request
Jul 25, 2025
Summary: X-link: pytorch/tnt#991 Context: DCP metadata collectives become prohibitively expensive as the job scale grows. This PR introduces rank-local 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. Test Plan: E2E UTs Differential Revision: D70112642
saumishr
force-pushed
the
export-D70112642
branch
from
b631684 to
3c6f8f9
Compare
saumishr
added a commit
to saumishr/tnt
that referenced
this pull request
Jul 25, 2025
Summary: X-link: pytorch/pytorch#147758 Context: DCP metadata collectives become prohibitively expensive as the job scale grows. This PR introduces rank-local 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: D70112642
|
This pull request was exported from Phabricator. Differential Revision: D70112642 |
pytorch-bot bot
pushed a commit
to pytorch/pytorch
that referenced
this pull request
Jul 30, 2025
Summary: X-link: pytorch/tnt#991 Context: DCP metadata collectives become prohibitively expensive as the job scale grows. This PR introduces rank-local 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. Test Plan: E2E UTs Differential Revision: D70112642
saumishr
added a commit
to saumishr/tnt
that referenced
this pull request
Jul 31, 2025
Summary: X-link: pytorch/pytorch#147758 Context: DCP metadata collectives become prohibitively expensive as the job scale grows. This PR introduces rank-local 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: D70112642
saumishr
force-pushed
the
export-D70112642
branch
from
0a24898 to
0229225
Compare
|
This pull request was exported from Phabricator. Differential Revision: D70112642 |
saumishr
added a commit
to saumishr/tnt
that referenced
this pull request
Jul 31, 2025
Summary: X-link: pytorch/pytorch#147758 Context: DCP metadata collectives become prohibitively expensive as the job scale grows. This PR introduces rank-local 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: D70112642
saumishr
force-pushed
the
export-D70112642
branch
from
0229225 to
c8ef529
Compare
|
This pull request was exported from Phabricator. Differential Revision: D70112642 |
saumishr
added a commit
to saumishr/pytorch
that referenced
this pull request
Jul 31, 2025
…ch#147758) Summary: X-link: pytorch/tnt#991 Context: DCP metadata collectives become prohibitively expensive as the job scale grows. This PR introduces rank-local 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. Test Plan: E2E UTs Differential Revision: D70112642
saumishr
added a commit
to saumishr/tnt
that referenced
this pull request
Jul 31, 2025
Summary: X-link: pytorch/pytorch#147758 Context: DCP metadata collectives become prohibitively expensive as the job scale grows. This PR introduces rank-local 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: D70112642
saumishr
force-pushed
the
export-D70112642
branch
from
c8ef529 to
d78a189
Compare
|
This pull request was exported from Phabricator. Differential Revision: D70112642 |
pytorch-bot bot
pushed a commit
to pytorch/pytorch
that referenced
this pull request
Aug 12, 2025
Summary: X-link: pytorch/tnt#991 Context: DCP metadata collectives become prohibitively expensive as the job scale grows. This PR introduces rank-local 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. Test Plan: E2E UTs Save and load test with internal DCP components: https://www.internalfb.com/mlhub/pipelines/runs/mast/torchx-textray-pretrain_mlm-lv5d7qcfmnqzkd Save and load test with OSS DCP components: https://www.internalfb.com/mlhub/pipelines/runs/mast/torchx-textray-pretrain_mlm-z1vz46vkkgtcld https://www.internalfb.com/mlhub/pipelines/runs/mast/torchx-textray-pretrain_mlm-njvvbn07rv5ckd Reviewed By: meetv18 Differential Revision: D70112642
saumishr
force-pushed
the
export-D70112642
branch
from
d78a189 to
7431191
Compare
saumishr
added a commit
to saumishr/tnt
that referenced
this pull request
Aug 12, 2025
Summary: X-link: pytorch/pytorch#147758 Context: DCP metadata collectives become prohibitively expensive as the job scale grows. This PR introduces rank-local 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. Reviewed By: meetv18 Differential Revision: D70112642
|
This pull request was exported from Phabricator. Differential Revision: D70112642 |
saumishr
added a commit
to saumishr/pytorch
that referenced
this pull request
Aug 13, 2025
…ch#147758) Summary: X-link: pytorch/tnt#991 Context: DCP metadata collectives become prohibitively expensive as the job scale grows. This PR introduces rank-local 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. Test Plan: E2E UTs Save and load test with internal DCP components: https://www.internalfb.com/mlhub/pipelines/runs/mast/torchx-textray-pretrain_mlm-lv5d7qcfmnqzkd Save and load test with OSS DCP components: https://www.internalfb.com/mlhub/pipelines/runs/mast/torchx-textray-pretrain_mlm-z1vz46vkkgtcld https://www.internalfb.com/mlhub/pipelines/runs/mast/torchx-textray-pretrain_mlm-njvvbn07rv5ckd Reviewed By: meetv18 Differential Revision: D70112642
saumishr
added a commit
to saumishr/tnt
that referenced
this pull request
Aug 13, 2025
Summary: X-link: pytorch/pytorch#147758 Context: DCP metadata collectives become prohibitively expensive as the job scale grows. This PR introduces rank-local 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. Reviewed By: meetv18 Differential Revision: D70112642
saumishr
force-pushed
the
export-D70112642
branch
from
7431191 to
c0a2e1c
Compare
|
This pull request was exported from Phabricator. Differential Revision: D70112642 |
saumishr
added a commit
to saumishr/pytorch
that referenced
this pull request
Aug 13, 2025
…ch#147758) Summary: X-link: pytorch/tnt#991 Context: DCP metadata collectives become prohibitively expensive as the job scale grows. This PR introduces rank-local 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. Test Plan: E2E UTs Save and load test with internal DCP components: https://www.internalfb.com/mlhub/pipelines/runs/mast/torchx-textray-pretrain_mlm-lv5d7qcfmnqzkd Save and load test with OSS DCP components: https://www.internalfb.com/mlhub/pipelines/runs/mast/torchx-textray-pretrain_mlm-z1vz46vkkgtcld https://www.internalfb.com/mlhub/pipelines/runs/mast/torchx-textray-pretrain_mlm-njvvbn07rv5ckd Reviewed By: meetv18 Differential Revision: D70112642
saumishr
added a commit
to saumishr/tnt
that referenced
this pull request
Aug 13, 2025
Summary: X-link: pytorch/pytorch#147758 Context: DCP metadata collectives become prohibitively expensive as the job scale grows. This PR introduces rank-local 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. Reviewed By: meetv18 Differential Revision: D70112642
saumishr
force-pushed
the
export-D70112642
branch
from
c0a2e1c to
c67c638
Compare
|
This pull request was exported from Phabricator. Differential Revision: D70112642 |
saumishr
added a commit
to saumishr/pytorch
that referenced
this pull request
Aug 13, 2025
…ch#147758) Summary: X-link: pytorch/tnt#991 Context: DCP metadata collectives become prohibitively expensive as the job scale grows. This PR introduces rank-local 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. Test Plan: E2E UTs Save and load test with internal DCP components: https://www.internalfb.com/mlhub/pipelines/runs/mast/torchx-textray-pretrain_mlm-lv5d7qcfmnqzkd Save and load test with OSS DCP components: https://www.internalfb.com/mlhub/pipelines/runs/mast/torchx-textray-pretrain_mlm-z1vz46vkkgtcld https://www.internalfb.com/mlhub/pipelines/runs/mast/torchx-textray-pretrain_mlm-njvvbn07rv5ckd Reviewed By: meetv18 Differential Revision: D70112642
saumishr
added a commit
to saumishr/tnt
that referenced
this pull request
Aug 13, 2025
Summary: X-link: pytorch/pytorch#147758 Context: DCP metadata collectives become prohibitively expensive as the job scale grows. This PR introduces rank-local 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. Reviewed By: meetv18 Differential Revision: D70112642
saumishr
force-pushed
the
export-D70112642
branch
from
c67c638 to
f82b20d
Compare
|
This pull request was exported from Phabricator. Differential Revision: D70112642 |
saumishr
added a commit
to saumishr/pytorch
that referenced
this pull request
Aug 13, 2025
…ch#147758) Summary: X-link: pytorch/tnt#991 Context: DCP metadata collectives become prohibitively expensive as the job scale grows. This PR introduces rank-local 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. Test Plan: E2E UTs Save and load test with internal DCP components: https://www.internalfb.com/mlhub/pipelines/runs/mast/torchx-textray-pretrain_mlm-lv5d7qcfmnqzkd Save and load test with OSS DCP components: https://www.internalfb.com/mlhub/pipelines/runs/mast/torchx-textray-pretrain_mlm-z1vz46vkkgtcld https://www.internalfb.com/mlhub/pipelines/runs/mast/torchx-textray-pretrain_mlm-njvvbn07rv5ckd Reviewed By: meetv18 Differential Revision: D70112642
Summary: X-link: pytorch/pytorch#147758 Context: DCP metadata collectives become prohibitively expensive as the job scale grows. This PR introduces rank-local 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. Reviewed By: meetv18 Differential Revision: D70112642
saumishr
force-pushed
the
export-D70112642
branch
from
f82b20d to
6d0e411
Compare
|
This pull request was exported from Phabricator. Differential Revision: D70112642 |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Summary:
X-link: pytorch/pytorch#147758
Context:
DCP metadata collectives become prohibitively expensive as the job scale grows. This PR introduces rank-local 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: D70112642