[spark] Add scan.max.records.per.partition config to split log table input partitions#3260
Open
Yohahaha wants to merge 1 commit intoapache:mainfrom
Open
[spark] Add scan.max.records.per.partition config to split log table input partitions#3260Yohahaha wants to merge 1 commit intoapache:mainfrom
Yohahaha wants to merge 1 commit intoapache:mainfrom
Conversation
Contributor
Author
Contributor
Author
|
@luoyuxia @fresh-borzoni PTAL! |
Member
fresh-borzoni
left a comment
fresh-borzoni
left a comment
There was a problem hiding this comment.
@Yohahaha Ty for the PR, overall LGTM, left minor comments, PTAL
| stopOffset: Long, | ||
| maxRecords: Long): Seq[InputPartition] = { | ||
| if ( | ||
| startOffset < 0 || stopOffset <= startOffset || stopOffset <= (startOffset + maxRecords) |
Member
There was a problem hiding this comment.
for the earliest mode we have sentinel -2L, I think it would result in a bug here, since we clamp to 1 split
| public class BucketOffsetsRetrieverImpl implements OffsetsInitializer.BucketOffsetsRetriever { | ||
| private final Admin flussAdmin; | ||
| private final TablePath tablePath; | ||
| private final Boolean fetchEarliestOffset; |
Member
There was a problem hiding this comment.
mb better to use primitive, since it doesn't allow null
| ) { | ||
| return Seq( | ||
| FlussAppendInputPartition(tableBucket, startOffset, stopOffset) | ||
| .asInstanceOf[InputPartition]) |
Member
There was a problem hiding this comment.
nit: I think it's redundant
| .map { | ||
| from => | ||
| FlussAppendInputPartition(tableBucket, from, math.min(from + step, stopOffset)) | ||
| .asInstanceOf[InputPartition] |
| } | ||
| } | ||
|
|
||
| test("Spark Read: split partition by config") { |
Member
There was a problem hiding this comment.
The test only checks row order/values, not partition count. Also what about partitioned tables? Earliest mode?
| .asInstanceOf[InputPartition] | ||
| Seq( | ||
| FlussAppendInputPartition(tableBucket, startOffset, stopOffset) | ||
| .asInstanceOf[InputPartition]) |
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
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 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.
Purpose
Linked issue: close #3215
Brief change log
scan.max.records.per.partitionconfig option for Spark log table reads. When set, each Flussbucket whose offset range exceeds this value will be split into multiple Spark input partitions, improving
read parallelism for large offset ranges.
BucketOffsetsRetrieverImplto support fetching real earliest offsets when needed.Tests
SparkLogTableReadTest: "Spark Read: split partition by config"API and Format
Documentation