Fix missing input bytes read metric for Parquet #1205
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: Jason Lowe <jlowe@nvidia.com>
|
build |
tgravescs
approved these changes
Nov 25, 2020
|
build |
kuhushukla
pushed a commit
to kuhushukla/spark-rapids
that referenced
this pull request
Nov 30, 2020
Signed-off-by: Jason Lowe <jlowe@nvidia.com>
nartal1
pushed a commit
to nartal1/spark-rapids
that referenced
this pull request
Jun 9, 2021
Signed-off-by: Jason Lowe <jlowe@nvidia.com>
nartal1
pushed a commit
to nartal1/spark-rapids
that referenced
this pull request
Jun 9, 2021
Signed-off-by: Jason Lowe <jlowe@nvidia.com>
tgravescs
pushed a commit
to tgravescs/spark-rapids
that referenced
this pull request
Nov 30, 2023
…IDIA#1205) Signed-off-by: spark-rapids automation <70000568+nvauto@users.noreply.github.com>
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Outdated suggestions cannot be applied.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.
Fixes #1199.
DataSourceRDDblindly assumes that any input occurs on the main task thread, but the plugin's Parquet readers support reading the data using a thread pool. WhenDataSourceRDDtries to grab the Hadoop FileSystem statistics for the main task thread, it finds no bytes read and unconditionally smashes the input metric to zero.Therefore this switches away from using
DataSourceRDDas-is and instead creates a derivation that performs everything it used to do except updating the input bytes metric. It is now up to the various scans (CSV, ORC, Parquet, etc.) to update the input bytes statistic. A wrapper class for partitioned readers was added to make it easy for inputs that don't yet support multithreaded reads (CSV and ORC), and the Parquet multithreaded readers were updated to retrieve the bytes read stats for each thread and report that back with the work being performed.