[SPARK-20952] ParquetFileFormat should forward TaskContext to its forkjoinpool #18176
Conversation
This file contains hidden or 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
Contributor
|
Jenkins this is ok to test |
|
Test build #77681 has finished for PR 18176 at commit
|
robert3005
force-pushed
the
rk/inheritable-thread-local
branch
from
06405ef to
6430dca
Compare
|
Test build #77753 has finished for PR 18176 at commit
|
Member
|
cc @JoshRosen who had some concerns with shared threadpool objects |
Contributor
|
Discussion on this is happening on JIRA: https://issues.apache.org/jira/browse/SPARK-20952 |
robert3005
force-pushed
the
rk/inheritable-thread-local
branch
from
6430dca to
25e8408
Compare
|
Test build #78531 has finished for PR 18176 at commit
|
Contributor
|
Jenkins this is ok to test |
|
Test build #78719 has finished for PR 18176 at commit
|
robert3005
changed the title
|
Can one of the admins verify this patch? |
rshkv
added a commit
to palantir/spark
that referenced
this pull request
Feb 23, 2021
…kjoinpool See ticket [1] and the PR [2] @robert3005 opened on apache/spark. It seems upstream wasn't convinced by our use-case. We need this change because Spark reads parquet footers on a different thread. Without this change, that thread doesn't inherit the thread-local that stores the TaskContext, meaning we don't have access to the properties stored inside that TaskContext Our internal filesystem needs those properties in the task context and fails to read footers without. [1] https://issues.apache.org/jira/browse/SPARK-20952 [2] apache#18176 Co-authored-by: Robert Kruszewski <[email protected]> Co-authored-by: Josh Casale <[email protected]> Co-authored-by: Will Raschkowski <[email protected]>
rshkv
added a commit
to palantir/spark
that referenced
this pull request
Feb 25, 2021
…kjoinpool See ticket [1] and the PR [2] @robert3005 opened on apache/spark. It seems upstream wasn't convinced by our use-case. We need this change because Spark reads parquet footers on a different thread. Without this change, that thread doesn't inherit the thread-local that stores the TaskContext, meaning we don't have access to the properties stored inside that TaskContext Our internal filesystem needs those properties in the task context and fails to read footers without. [1] https://issues.apache.org/jira/browse/SPARK-20952 [2] apache#18176 Co-authored-by: Robert Kruszewski <[email protected]> Co-authored-by: Josh Casale <[email protected]> Co-authored-by: Will Raschkowski <[email protected]>
rshkv
added a commit
to palantir/spark
that referenced
this pull request
Feb 26, 2021
…kjoinpool See ticket [1] and the PR [2] @robert3005 opened on apache/spark. It seems upstream wasn't convinced by our use-case. We need this change because Spark reads parquet footers on a different thread. Without this change, that thread doesn't inherit the thread-local that stores the TaskContext, meaning we don't have access to the properties stored inside that TaskContext Our internal filesystem needs those properties in the task context and fails to read footers without. [1] https://issues.apache.org/jira/browse/SPARK-20952 [2] apache#18176 Co-authored-by: Robert Kruszewski <[email protected]> Co-authored-by: Josh Casale <[email protected]> Co-authored-by: Will Raschkowski <[email protected]>
rshkv
added a commit
to palantir/spark
that referenced
this pull request
Feb 26, 2021
…kjoinpool See ticket [1] and the PR [2] @robert3005 opened on apache/spark. It seems upstream wasn't convinced by our use-case. We need this change because Spark reads parquet footers on a different thread. Without this change, that thread doesn't inherit the thread-local that stores the TaskContext, meaning we don't have access to the properties stored inside that TaskContext Our internal filesystem needs those properties in the task context and fails to read footers without. [1] https://issues.apache.org/jira/browse/SPARK-20952 [2] apache#18176 Co-authored-by: Robert Kruszewski <[email protected]> Co-authored-by: Josh Casale <[email protected]> Co-authored-by: Will Raschkowski <[email protected]>
jdcasale
added a commit
to palantir/spark
that referenced
this pull request
Mar 3, 2021
…kjoinpool See ticket [1] and the PR [2] @robert3005 opened on apache/spark. It seems upstream wasn't convinced by our use-case. We need this change because Spark reads parquet footers on a different thread. Without this change, that thread doesn't inherit the thread-local that stores the TaskContext, meaning we don't have access to the properties stored inside that TaskContext Our internal filesystem needs those properties in the task context and fails to read footers without. [1] https://issues.apache.org/jira/browse/SPARK-20952 [2] apache#18176 Co-authored-by: Robert Kruszewski <[email protected]> Co-authored-by: Josh Casale <[email protected]> Co-authored-by: Will Raschkowski <[email protected]>
rshkv
added a commit
to palantir/spark
that referenced
this pull request
Mar 4, 2021
…kjoinpool See ticket [1] and the PR [2] @robert3005 opened on apache/spark. It seems upstream wasn't convinced by our use-case. We need this change because Spark reads parquet footers on a different thread. Without this change, that thread doesn't inherit the thread-local that stores the TaskContext, meaning we don't have access to the properties stored inside that TaskContext Our internal filesystem needs those properties in the task context and fails to read footers without. [1] https://issues.apache.org/jira/browse/SPARK-20952 [2] apache#18176 Co-authored-by: Robert Kruszewski <[email protected]> Co-authored-by: Josh Casale <[email protected]> Co-authored-by: Will Raschkowski <[email protected]>
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
6 participants
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.
What changes were proposed in this pull request?
Make TaskContext reference an InheritableTheadLocal so thread pools spun up inside tasks have access to the reference
How was this patch tested?
Added tests