Add partitioning support in parquet writer #9810
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
…en a previour write(table) call failed
github-actions
bot
added
Python
Exception raised by writer needs to be same as that raised by pandas. If user_data is constructed earlier using pyarrow then the exception is raised early and is different
devavret
added
4 - Needs cuIO Reviewer
breaking
|
Please add description to the PR |
vuule
requested changes
Dec 13, 2021
galipremsagar
approved these changes
Dec 14, 2021
|
@gpucibot merge |
bdice
reviewed
Dec 15, 2021
| * @return this for chaining. | ||
| */ | ||
| parquet_writer_options_builder& column_chunks_file_path(std::string file_path) | ||
| parquet_writer_options_builder& column_chunks_file_paths(std::vector<std::string> file_paths) |
There was a problem hiding this comment.
This change breaks compilation of benchmarks, which were not updated in this PR:
We should ensure that CI compiles the benchmarks even if they don't run.
There was a problem hiding this comment.
I have the fix in #9776, just needs a small donation of one code review :P
rapids-bot bot
pushed a commit
that referenced
this pull request
Dec 15, 2021
This fixes a compilation error introduced in #9810. Tagging @devavret @vuule for review. Feel free to push to this PR with any fixes. Authors: - Bradley Dice (https://github.com/bdice) Approvers: - Vukasin Milovanovic (https://github.com/vuule) - Nghia Truong (https://github.com/ttnghia) URL: #9905
|
It also would have been nice to get a heads up that the java build broke |
Sorry about that 😞 . I presumed that spark team must've been notified if a PR had a |
rapids-bot bot
pushed a commit
that referenced
this pull request
Dec 15, 2021
This fixes the java build after #9810 went in. There is a lot of copy/paste in this first draft, because I just wanted to get something to work. Not sure if it is worth going back to make it common everywhere. Authors: - Robert (Bobby) Evans (https://github.com/revans2) Approvers: - Jason Lowe (https://github.com/jlowe) URL: #9908
rapids-bot bot
pushed a commit
that referenced
this pull request
Jan 10, 2022
Makes use of the efficient partitioned writing support added in #9810 to improve performance of partitioned parquet dataset writing. Closes #5059 Authors: - Devavret Makkar (https://github.com/devavret) Approvers: - Vukasin Milovanovic (https://github.com/vuule) - GALI PREM SAGAR (https://github.com/galipremsagar) - Richard (Rick) Zamora (https://github.com/rjzamora) URL: #9971
vyasr
added
4 - Needs Review
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.
Contributes to #5059
Adds libcudf support for writing partitioned datasets in parquet writer. With the new API, one can specify a vector of
{start_row, num_rows}structs along with a table st slices of the input table gets written to the corresponding sink.Adds Multi-sink support in
sink_info