Skip to content

Pull requests: NVIDIA/spark-rapids

New pull request New
31 Open 7,431 Closed
31 Open 7,431 Closed
Author
Filter by author
Loading
Label
Filter by label
Loading
Use alt + click/return to exclude labels
or ⇧ + click/return for logical OR
Projects
Filter by project
Loading
Milestones
Filter by milestone
Loading
Reviews
Filter by reviews
No reviews Review required Approved review Changes requested
Assignee
Filter by who’s assigned
Assigned to nobody Loading
Sort
Sort by
Newest Oldest Most commented Least commented Recently updated Least recently updated Best match
Most reactions
👍 👎 😄 🎉 😕 ❤️ 🚀 👀

Pull requests list

[AutoSparkUT]Enable RapidsCsvExpressionsSuite & RapidsCSVInferSchemaSuite test Only impacts tests
#13911 opened Dec 1, 2025 by GaryShen2008 Loading…
4
Refine GpuTaskMetrics over SpillFrameWork
#13905 opened Nov 28, 2025 by sperlingxx Loading…
1
19
Fix join bug on csv datasources bug Something isn't working
#13903 opened Nov 28, 2025 by res-life Loading…
1
5
Enable iceberg write by default, and disable dml iceberg operation on mor table by default.
#13891 opened Nov 27, 2025 by liurenjie1024 Loading…
3 tasks done
5
Add in API definitions for Rapids UDAF
#13870 opened Nov 25, 2025 by firestarman Loading…
13
Replace apache.commons.lang3.StringUtils.abbreviate calls improve
#13863 opened Nov 25, 2025 by mythrocks Loading…
@mythrocks
7
Add a non strict mode for lore dump task Work required that improves the product but is not user facing
#13845 opened Nov 21, 2025 by thirtiseven Loading…
2 of 3 tasks
1
@thirtiseven
13
Simplify gpu transform operations in iceberg task Work required that improves the product but is not user facing
#13844 opened Nov 21, 2025 by liurenjie1024 • Draft
2 of 3 tasks
1
3
Make convert in R2C retryable to prevent Host OOM reliability Features to improve reliability or bugs that severly impact the reliability of the plugin
#13842 opened Nov 21, 2025 by thirtiseven Loading…
1 of 3 tasks
16
Add coalescing support to multithreaded reader when reading Deletion Vectors task Work required that improves the product but is not user facing
#13783 opened Nov 14, 2025 by razajafri • Draft
1 of 3 tasks
1
4
Kudo gpu cpu fallback
#13781 opened Nov 13, 2025 by zpuller • Draft
3 tasks
3
Add support for Deletion Vectors to MultiFileParquetPartitionReader feature request New feature or request
#13744 opened Nov 11, 2025 by razajafri Loading…
2 of 3 tasks
1
11
[DO NOT REVIEW] rapids shuffle manager writer use as much memory as allowed
#13724 opened Nov 6, 2025 by binmahone • Draft
@binmahone
6
enable kudo gpu write and read by default
#13720 opened Nov 5, 2025 by zpuller • Draft
3 tasks
6
fix ucx init shuffle manager for spark 4 bug Something isn't working
#13665 opened Oct 27, 2025 by zpuller • Draft
3 tasks
1
9
Add in support for sort merge join optimization performance A performance related task/issue
#13640 opened Oct 21, 2025 by revans2 • Draft
6
Remove LazySpillableColumnarBatch from join code performance A performance related task/issue
#13625 opened Oct 20, 2025 by abellina Loading…
2 of 3 tasks
18
Reimplement min_by and max_by aggregations improve performance A performance related task/issue SQL part of the SQL/Dataframe plugin
#13603 opened Oct 13, 2025 by ttnghia Loading…
@ttnghia
4
[DOC]Add user doc for RAPIDS UDAF [skip ci] documentation Improvements or additions to documentation
#13602 opened Oct 13, 2025 by firestarman • Draft
1
4
reduce NDS benchmark e2e time by optimizing RapidsShuffleManager performance A performance related task/issue
#13479 opened Sep 24, 2025 by binmahone Loading…
@binmahone
80
Support UDAF on GPU[databricks] feature request New feature or request performance A performance related task/issue
#13450 opened Sep 19, 2025 by firestarman • Draft
2
101
Stronger typing for GpuV1FallbackWriters [databricks] task Work required that improves the product but is not user facing
#13426 opened Sep 5, 2025 by jihoonson Loading…
2 of 3 tasks
1
7
Unshimmify Arm helper methods to expose as public API task Work required that improves the product but is not user facing
#13424 opened Sep 5, 2025 by rishic3 • Draft
1 of 3 tasks
1
@rishic3
13
collecting gpu metrics with NVML
#13398 opened Sep 2, 2025 by binmahone Loading…
@binmahone
14
Adds in support for a "CpuBridge" that lets us fall back to the CPU on a per-expression level instead of per-SparkPlan node level. [databricks] performance A performance related task/issue
#13368 opened Aug 27, 2025 by revans2 Loading…
2 of 3 tasks
59
ProTip! no:milestone will show everything without a milestone.