[tune] Add multinode sync test #23229
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
added 4 commits
March 15, 2022 19:52
# Conflicts: # python/ray/autoscaler/_private/fake_multi_node/docker_monitor.py # python/ray/autoscaler/_private/fake_multi_node/node_provider.py # python/ray/autoscaler/_private/fake_multi_node/test_utils.py
xwjiang2010
reviewed
Mar 17, 2022
| @@ -798,6 +798,9 @@ def restore(self, trial) -> None: | |||
| else: | |||
| logger.debug("Trial %s: Attempting restore from %s", trial, value) | |||
| if trial.uses_cloud_checkpointing or not trial.sync_on_checkpoint: | |||
| # If using cloud checkpointing, trial will get cp from cloud. | |||
| # If not syncing to driver, assume it has access to the cp | |||
There was a problem hiding this comment.
and what guarantees that this is the case?
There was a problem hiding this comment.
Nothing (in fact we sometimes run into errors when this is misconfigured). I've added the comment for future reference (as I find this part confusing). We should probably have better guarantees in the future.
|
Overall looks good, thanks Kai! Just some comments on documentation for future maintenance of these tests. |
xwjiang2010
approved these changes
Mar 21, 2022
| @@ -158,6 +158,16 @@ some limitations still remain: | |||
|
|
|||
| 3. In docker-in-docker setups, a careful setup has to be followed to make the fake multinode docker provider work (see below). | |||
|
|
|||
| Shared directories within the docker environment | |||
There was a problem hiding this comment.
Thanks for adding this!
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Outdated suggestions cannot be applied.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 adds a multinode checkpoint/restore test for Ray Tune. This covers some of the functionality of the release tests, but in a more controlled environment. In a follow-up PR, we should test (mocked) cloud checkpointing, too.
Related issue number
Checks
scripts/format.shto lint the changes in this PR.