[jira] [Commented] (FLINK-8739) Optimize runtime support for distinct filter
[ https://issues.apache.org/jira/browse/FLINK-8739?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16718569#comment-16718569 ] ASF GitHub Bot commented on FLINK-8739: --- dianfu opened a new pull request #7286: [FLINK-8739] [table] Optimize DISTINCE aggregates to use the same distinct accumulator if possible URL: https://github.com/apache/flink/pull/7286 ## What is the purpose of the change *This pull request optimizes the DISTINCT aggregates such as `SELECT COUNT(DISTINCT(a)), SUM(DISTINCT(a))` to use the same distinct accumulator for performance optimization.* ## Brief change log - *Separate DistinctAccumulator and the actual accumulator* - *Optimize the AggregateCodeGeneration to reuse the same DistinctAccumulator if possible* ## Verifying this change This change is already covered by existing tests, such as *SqlITCase#testDistinctAggOnRowTimeTumbleWindow, AggregateITCase, OverWindowITCase*. Still missing: For the harness test, I will add more harness tests base on #7253 ## Does this pull request potentially affect one of the following parts: - Dependencies (does it add or upgrade a dependency): (yes / no) - The public API, i.e., is any changed class annotated with `@Public(Evolving)`: (yes / no) - The serializers: (yes / no / don't know) - The runtime per-record code paths (performance sensitive): (yes / no / don't know) - Anything that affects deployment or recovery: JobManager (and its components), Checkpointing, Yarn/Mesos, ZooKeeper: (yes / no / don't know) - The S3 file system connector: (yes / no / don't know) ## Documentation - Does this pull request introduce a new feature? (yes / no) - If yes, how is the feature documented? (not applicable / docs / JavaDocs / not documented) This is an automated message from the Apache Git Service. To respond to the message, please log on GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: us...@infra.apache.org > Optimize runtime support for distinct filter > > > Key: FLINK-8739 > URL: https://issues.apache.org/jira/browse/FLINK-8739 > Project: Flink > Issue Type: Improvement > Components: Table API SQL >Reporter: Rong Rong >Assignee: Dian Fu >Priority: Major > Labels: pull-request-available > > Possible optimizaitons: > 1. Decouple distinct map and actual accumulator so that they can separately > be created in codegen. > 2. Reuse same distinct accumulator for filtering, e.g. `SELECT > COUNT(DISTINCT(a)), SUM(DISTINCT(a))` should reuse the same distinct map. -- This message was sent by Atlassian JIRA (v7.6.3#76005)
[jira] [Commented] (FLINK-8739) Optimize runtime support for distinct filter
[ https://issues.apache.org/jira/browse/FLINK-8739?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16706595#comment-16706595 ] Dian Fu commented on FLINK-8739: [~walterddr] Thanks a lot. I'll provide a PR ASAP and looking forward to your review. > Optimize runtime support for distinct filter > > > Key: FLINK-8739 > URL: https://issues.apache.org/jira/browse/FLINK-8739 > Project: Flink > Issue Type: Improvement > Components: Table API SQL >Reporter: Rong Rong >Priority: Major > > Possible optimizaitons: > 1. Decouple distinct map and actual accumulator so that they can separately > be created in codegen. > 2. Reuse same distinct accumulator for filtering, e.g. `SELECT > COUNT(DISTINCT(a)), SUM(DISTINCT(a))` should reuse the same distinct map. -- This message was sent by Atlassian JIRA (v7.6.3#76005)
[jira] [Commented] (FLINK-8739) Optimize runtime support for distinct filter
[ https://issues.apache.org/jira/browse/FLINK-8739?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16704949#comment-16704949 ] Rong Rong commented on FLINK-8739: -- No I haven't started. please feel free to assign to yourself. > Optimize runtime support for distinct filter > > > Key: FLINK-8739 > URL: https://issues.apache.org/jira/browse/FLINK-8739 > Project: Flink > Issue Type: Improvement > Components: Table API SQL >Reporter: Rong Rong >Priority: Major > > Possible optimizaitons: > 1. Decouple distinct map and actual accumulator so that they can separately > be created in codegen. > 2. Reuse same distinct accumulator for filtering, e.g. `SELECT > COUNT(DISTINCT(a)), SUM(DISTINCT(a))` should reuse the same distinct map. -- This message was sent by Atlassian JIRA (v7.6.3#76005)
[jira] [Commented] (FLINK-8739) Optimize runtime support for distinct filter
[ https://issues.apache.org/jira/browse/FLINK-8739?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16704356#comment-16704356 ] Dian Fu commented on FLINK-8739: Hi [~walterddr], are you still working on this ticket? I'd like to help on this ticket if you haven't started the work. Thanks a lot. > Optimize runtime support for distinct filter > > > Key: FLINK-8739 > URL: https://issues.apache.org/jira/browse/FLINK-8739 > Project: Flink > Issue Type: Improvement > Components: Table API SQL >Reporter: Rong Rong >Priority: Major > > Possible optimizaitons: > 1. Decouple distinct map and actual accumulator so that they can separately > be created in codegen. > 2. Reuse same distinct accumulator for filtering, e.g. `SELECT > COUNT(DISTINCT(a)), SUM(DISTINCT(a))` should reuse the same distinct map. -- This message was sent by Atlassian JIRA (v7.6.3#76005)
