Add in retry for ORC writes [databricks] #7972
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
Signed-off-by: Robert (Bobby) Evans <[email protected]>
|
build |
jlowe
reviewed
Mar 29, 2023
sql-plugin/src/main/scala/com/nvidia/spark/rapids/ColumnarOutputWriter.scala
Outdated
Show resolved
Hide resolved
sql-plugin/src/main/scala/org/apache/spark/sql/rapids/GpuOrcFileFormat.scala
Outdated
Show resolved
Hide resolved
sql-plugin/src/main/scala/com/nvidia/spark/rapids/GpuTransitionOverrides.scala
Outdated
Show resolved
Hide resolved
sql-plugin/src/main/scala/com/nvidia/spark/rapids/GpuParquetFileFormat.scala
Outdated
Show resolved
Hide resolved
revans2
changed the title
|
build |
|
build |
jlowe
previously approved these changes
Mar 30, 2023
|
Looks like the OOM injection PR messed with some things for this PR, so I will spend some time to debug this... |
|
build |
abellina
approved these changes
Mar 30, 2023
jlowe
approved these changes
Mar 30, 2023
sameerz
added
the
reliability
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Outdated suggestions cannot be applied.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.
This fixes #7341
This fixes #7960
I also included a fix for metrics because for some reason when writing data the metrics were not deserialized until all of the query was done. For now I just made the task level metrics go on most GPU operators.
I did some performance testing with smaller and smaller memory. The issues that I was seeing were mostly with not being able to split inputs on a parquet read (when I got down to 4 GiB of GPU memory and 4x parallelism) I also saw some issues with running out of memory when trying to read back in spilled data.
It is a nice logarithmic looking performance drop off, which is nice to see.