Write metadata file #864
Merged
Write metadata file #864
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
RobbeSneyders
added a commit
that referenced
this pull request
Feb 24, 2024
We reintroduced writing the metadata file in #864 to preserve the divisions of the data when writing and reading again. We turned this behavior off in the past, but without proper documentation of the reason. I'm now running into issues with Dask workers dying when writing large datasets though, presumably because of the metadata file, as documented in these Dask issues: - dask/dask#6600 - dask/dask#3873 - dask/dask#8901 Also, while I ran into issues with the preservation of divisions before, I can't reproduce this locally with a small example. Let's turn writing metadata off again and validate if we are still having issues with 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Outdated suggestions cannot be applied.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.
We currently don't preserve the divisions of the data when writing and reading again, which leads to errors when merging datasets with a low and high amount of partitions. This PR enables the writing of a metadata file which should fix this.
This was originally introduced in #391, which contains more information, but then later reverted in #403 without a clear reasoning.
Let's reactivate it, and if there's a reason to remove it again, let's document it properly.