Raise exception if dask cluster is incompatible with Dataset.to_ddf return type
#339
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
Documentation preview |
karlhigley
approved these changes
Jun 1, 2023
This was referenced Jun 1, 2023
Draft
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Outdated suggestions cannot be applied.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.
Raise exception if dask cluster is incompatible with
Dataset.to_ddfreturn type.This was motivated by a test we had in Merlin Systems where we were using the
Distributedhelper class with argumentcluster_type="cpu"and using aDataset(df).to_ddf()which returned adask_cudf.DataFrame. This type of DataFrame is incompatible with the cluster typedistributed.LocalCluster, since it doesn't handle gpu-backed dataframes. When trying to run a dask computation in this scenario, unexpected errors can happen related to unmanaged memory.This PR aims to reduce the risk of this scenario (for the case where
to_ddfis called in a context where the global dask client is defined). A longer-term fix is to add an error indask_cudfordistributedto check for this scenario and alert the user.