Minimize io operations in experiment, avoid persisting experiment to disk before run/dryrun#150
Merged
hemildesai merged 3 commits intomainfrom Feb 14, 2025
Merged
Minimize io operations in experiment, avoid persisting experiment to disk before run/dryrun#150hemildesai merged 3 commits intomainfrom
hemildesai merged 3 commits intomainfrom
Conversation
…disk before run/dryrun Signed-off-by: Hemil Desai <hemild@nvidia.com>
Signed-off-by: Hemil Desai <hemild@nvidia.com>
Kipok
approved these changes
Feb 13, 2025
Contributor
Kipok
left a comment
Kipok
left a comment
There was a problem hiding this comment.
I didn't test the speed, but job submission seems to work correctly
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
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
3 participants
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
This PR solves two problems:
Before, experiment and job config was being persisted to disk even if
exp.runwas not called. This created unnecessary directories and made it harder to track down experiments that actually ran. Now, it persists information only whenexp.runis called thus eliminating noise.Before,
_save_jobswas called after everyexp.addand after every job launch. This is unnecessary and increases io times for experiments that have thousands of jobs. Now we only call_save_jobstwice. Once at the start ofexp.runand once at the end. This will be followed by a PR that cancels all submitted jobs ifexp.runfails midway.