Increase maxLocalExchangeBufferSize to 128MB #18212
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
gaurav8297
force-pushed
the
higher_buffer_size_for_write
branch
from
ca39f8c to
812a42b
Compare
sopel39
reviewed
Jul 11, 2023
plugin/trino-hive/src/test/java/io/trino/plugin/hive/BaseHiveConnectorTest.java
Show resolved
Hide resolved
|
there is test failure. Is it relevant? |
It's this issue #18217 |
Closed
core/trino-main/src/main/java/io/trino/execution/TaskManagerConfig.java
Outdated
Show resolved
Hide resolved
sopel39
approved these changes
Jul 12, 2023
There was a problem hiding this comment.
lgtm % @raunaqmorarka comment. Maybe we can just increase local exchange buffer overall
This change will help increase the performance of writes when local exchange is used either because of partitioning the data or scaling the writers. Benchmarks for unpartitioned write: Before:ku Input Size: 6B rows Time: 4:46 mins Peak Mem: 59.9GB After: Input Size: 6B rows Time: 3:45 mins Peak Mem: 61.2GB Benchmarks for partitioned write: Before: Input Size: 3B rows (3 skewed partitions) Time: 2:33 mins Peak Mem: 89.8GB After: Input Size: 3B rows (3 skewed partitions) Time: 1:57 mins Peak Mem: 92.3GB
gaurav8297
force-pushed
the
higher_buffer_size_for_write
branch
from
812a42b to
77a5b90
Compare
|
I update the default value to 128MB as @raunaqmorarka mentioned. Also, here is a benchmark report for tpcds. |
gaurav8297
changed the title
raunaqmorarka
approved these changes
Jul 13, 2023
Closed
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Outdated suggestions cannot be applied.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 change will help increase the performance of writes when local exchange is used either due to partitioning the data or scaling the writers.
Benchmarks for unpartitioned write:
Before:ku
Input Size: 6B rows
Time: 4:46 mins
Peak Mem: 59.9GB
After:
Input Size: 6B rows
Time: 3:45 mins
Peak Mem: 61.2GB
Benchmarks for partitioned write:
Before:
Input Size: 3B rows (3 skewed partitions)
Time: 2:33 mins
Peak Mem: 89.8GB
After:
Input Size: 3B rows (3 skewed partitions)
Time: 1:57 mins
Peak Mem: 92.3GB
Description
Additional context and related issues
Release notes
( ) This is not user-visible or docs only and no release notes are required.
( ) Release notes are required, please propose a release note for me.
( ) Release notes are required, with the following suggested text: