[Train] (Bandaid) Mitigate OOMs on checkpointing #33089
Merged
Conversation
This file contains 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
Signed-off-by: Antoni Baum <antoni.baum@protonmail.com>
7 tasks
amogkam
reviewed
Mar 7, 2023
Signed-off-by: Antoni Baum <antoni.baum@protonmail.com>
amogkam
approved these changes
Mar 7, 2023
ProjectsByJackHe
pushed a commit
to ProjectsByJackHe/ray
that referenced
this pull request
Mar 21, 2023
* [Train] (Bandaid) Mitigate OOMs on checkpointing This PR introduces a tactical mitigation of ray-project#33073 by ensuring that the rank 0 worker is colocated with the Trainable if possible, allowing for lazy checkpointing to be used (and thus avoiding a situation where the entire checkpoint is loaded into memory to be passed to object store). THIS IS NOT A LONG TERM FIX. It is necessary to unblock LLM use cases for now. A proper solution should be implemented soon. --------- Signed-off-by: Antoni Baum <antoni.baum@protonmail.com> Signed-off-by: Jack He <jackhe2345@gmail.com>
edoakes
pushed a commit
to edoakes/ray
that referenced
this pull request
Mar 22, 2023
* [Train] (Bandaid) Mitigate OOMs on checkpointing This PR introduces a tactical mitigation of ray-project#33073 by ensuring that the rank 0 worker is colocated with the Trainable if possible, allowing for lazy checkpointing to be used (and thus avoiding a situation where the entire checkpoint is loaded into memory to be passed to object store). THIS IS NOT A LONG TERM FIX. It is necessary to unblock LLM use cases for now. A proper solution should be implemented soon. --------- Signed-off-by: Antoni Baum <antoni.baum@protonmail.com> Signed-off-by: Edward Oakes <ed.nmi.oakes@gmail.com>
peytondmurray
pushed a commit
to peytondmurray/ray
that referenced
this pull request
Mar 22, 2023
* [Train] (Bandaid) Mitigate OOMs on checkpointing This PR introduces a tactical mitigation of ray-project#33073 by ensuring that the rank 0 worker is colocated with the Trainable if possible, allowing for lazy checkpointing to be used (and thus avoiding a situation where the entire checkpoint is loaded into memory to be passed to object store). THIS IS NOT A LONG TERM FIX. It is necessary to unblock LLM use cases for now. A proper solution should be implemented soon. --------- Signed-off-by: Antoni Baum <antoni.baum@protonmail.com>
elliottower
pushed a commit
to elliottower/ray
that referenced
this pull request
Apr 22, 2023
* [Train] (Bandaid) Mitigate OOMs on checkpointing This PR introduces a tactical mitigation of ray-project#33073 by ensuring that the rank 0 worker is colocated with the Trainable if possible, allowing for lazy checkpointing to be used (and thus avoiding a situation where the entire checkpoint is loaded into memory to be passed to object store). THIS IS NOT A LONG TERM FIX. It is necessary to unblock LLM use cases for now. A proper solution should be implemented soon. --------- Signed-off-by: Antoni Baum <antoni.baum@protonmail.com> Signed-off-by: elliottower <elliot@elliottower.com>
ProjectsByJackHe
pushed a commit
to ProjectsByJackHe/ray
that referenced
this pull request
May 4, 2023
* [Train] (Bandaid) Mitigate OOMs on checkpointing This PR introduces a tactical mitigation of ray-project#33073 by ensuring that the rank 0 worker is colocated with the Trainable if possible, allowing for lazy checkpointing to be used (and thus avoiding a situation where the entire checkpoint is loaded into memory to be passed to object store). THIS IS NOT A LONG TERM FIX. It is necessary to unblock LLM use cases for now. A proper solution should be implemented soon. --------- Signed-off-by: Antoni Baum <antoni.baum@protonmail.com> Signed-off-by: Jack He <jackhe2345@gmail.com>
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.
Why are these changes needed?
This PR introduces a tactical mitigation of #33073 by ensuring that the rank 0 worker is colocated with the Trainable if possible, allowing for lazy checkpointing to be used (and thus avoiding a situation where the entire checkpoint is loaded into memory to be passed to object store).
THIS IS NOT A LONG TERM FIX. It is necessary to unblock LLM use cases for now. A proper solution should be implemented soon.
Related issue number
Checks
git commit -s) in this PR.scripts/format.shto lint the changes in this PR.