Set initial number of tasks for scaled writer with HBO by feilong-liu · Pull Request #20901 · prestodb/presto

feilong-liu · 2023-09-18T21:33:51Z

Description

Addresses #20355
Record the number of tasks used in scaled writers in HBO, and use HBO to set the initial number of writers to begin with for scaled writers.

Motivation and Context

Scaled writers first have only 1 task to write data out, and increase the number of tasks as needed when the source is throttled. In this PR, the scaled writer will start with a number based on the number of previous runs, so that it can have larger parallelism in the beginning and hence improve latency.

Impact

Latency improvement for scaled writer pipelines

Test Plan

Test query

Also run with verifier suite

Contributor checklist

Please make sure your submission complies with our development, formatting, commit message, and attribution guidelines.
PR description addresses the issue accurately and concisely. If the change is non-trivial, a GitHub Issue is referenced.
Documented new properties (with its default value), SQL syntax, functions, or other functionality.
If release notes are required, they follow the release notes guidelines.
Adequate tests were added if applicable.
CI passed.

Release Notes

Please follow release notes guidelines and fill in the release notes below.

== RELEASE NOTES ==

General Changes
* Add optimization for scaled writers with HBO, it can improve latency for query with scaled writer enabled. It's controlled by session property `enable_hbo_for_scaled_writer` and default to false.

feilong-liu · 2023-09-27T17:57:00Z

presto-main/src/main/java/com/facebook/presto/cost/HistoryBasedPlanStatisticsTracker.java

Get the number of tasks for the stage, and record it.

feilong-liu · 2023-09-27T18:04:07Z

presto-main/src/main/java/com/facebook/presto/cost/PlanNodeStatsEstimate.java

Add table writer stats to estimate

presto-main/src/main/java/com/facebook/presto/sql/planner/planPrinter/TextRenderer.java

feilong-liu · 2023-09-27T18:06:56Z

presto-spi/src/main/java/com/facebook/presto/spi/statistics/PlanStatistics.java

Add table writer node statistics to plan statistics

feilong-liu · 2023-09-27T18:09:45Z

presto-main/src/main/java/com/facebook/presto/execution/scheduler/ScaledWriterScheduler.java

Add a field which specify the number of tasks to start from for scaled writer

feilong-liu · 2023-09-27T18:10:21Z

presto-main/src/main/java/com/facebook/presto/execution/scheduler/ScaledWriterScheduler.java

Start from 1 if no initial task number specified

feilong-liu · 2023-09-27T19:19:01Z

presto-main/src/main/java/com/facebook/presto/sql/planner/PlanFragmenterUtils.java

Get the suggested number of table writer tasks from query plan, it finds the TableWriterNode, and read from its task number if scale writer field.

feilong-liu · 2023-09-27T19:19:34Z

presto-main/src/main/java/com/facebook/presto/sql/planner/PlanOptimizers.java

Add a rule to set the initial number of tasks for a table writer

presto-main/src/main/java/com/facebook/presto/sql/planner/planPrinter/PlanPrinter.java

feilong-liu · 2023-09-27T19:25:38Z

presto-main/src/main/java/com/facebook/presto/sql/planner/plan/TableWriterNode.java

Add a field to specify the number of tasks to begin with if it's a scaled writer

feilong-liu · 2023-09-27T22:31:37Z

presto-main/src/main/java/com/facebook/presto/execution/scheduler/SectionExecutionFactory.java

Get the preferred number of tasks from table writer nodes in the plan

mlyublena · 2023-09-29T00:00:17Z

presto-main/src/main/java/com/facebook/presto/cost/HistoryBasedPlanStatisticsTracker.java

is TableWriterMergeNode relevant here?

No, the scaled writer optimization is only related to table writer node, not related to table writer merger node.

presto-main/src/main/java/com/facebook/presto/sql/planner/iterative/rule/ScaledWriterRule.java

presto-main/src/main/java/com/facebook/presto/sql/planner/plan/TableWriterNode.java

presto-spi/src/main/java/com/facebook/presto/spi/statistics/TableWriterNodeStatistics.java

presto-main/src/main/java/com/facebook/presto/cost/TableWriterNodeStatsEstimate.java

presto-main/src/main/java/com/facebook/presto/execution/scheduler/ScaledWriterScheduler.java

presto-main/src/main/java/com/facebook/presto/sql/planner/planPrinter/PlanPrinter.java

presto-main/src/main/java/com/facebook/presto/sql/planner/planPrinter/TextRenderer.java

presto-main/src/main/java/com/facebook/presto/SystemSessionProperties.java

mlyublena · 2023-10-10T16:45:11Z

presto-main/src/main/java/com/facebook/presto/sql/planner/iterative/rule/ScaledWriterRule.java

I just merged #20990 that tracks which optimizers were cost-based and the source of stats used (CBO/HBO).
Can you override functions isCostBased and getStatsSource so this optimizer also gets tracked?

Added in a separate PR #21120

feilong-liu requested review from a team and shrinidhijoshi as code owners September 18, 2023 21:33

feilong-liu requested a review from presto-oss September 18, 2023 21:33

feilong-liu marked this pull request as draft September 18, 2023 21:33

feilong-liu force-pushed the scaledwriter_hbo branch 6 times, most recently from 3c0ba5b to f8efc49 Compare September 19, 2023 23:13

feilong-liu force-pushed the scaledwriter_hbo branch from f8efc49 to 214c542 Compare September 27, 2023 17:55

feilong-liu commented Sep 27, 2023

View reviewed changes

feilong-liu force-pushed the scaledwriter_hbo branch 4 times, most recently from 0660644 to 70ad759 Compare September 27, 2023 22:34

feilong-liu commented Sep 27, 2023

View reviewed changes

feilong-liu marked this pull request as ready for review September 28, 2023 00:06

feilong-liu requested review from kaikalur, mlyublena and pranjalssh September 28, 2023 00:06

mlyublena reviewed Sep 29, 2023

View reviewed changes

feilong-liu requested a review from mlyublena October 2, 2023 17:34

mlyublena reviewed Oct 2, 2023

View reviewed changes

presto-main/src/main/java/com/facebook/presto/sql/planner/iterative/rule/ScaledWriterRule.java Outdated Show resolved Hide resolved

mlyublena reviewed Oct 2, 2023

View reviewed changes

presto-main/src/main/java/com/facebook/presto/sql/planner/plan/TableWriterNode.java Outdated Show resolved Hide resolved

mlyublena reviewed Oct 2, 2023

View reviewed changes

presto-spi/src/main/java/com/facebook/presto/spi/statistics/TableWriterNodeStatistics.java Outdated Show resolved Hide resolved

mlyublena approved these changes Oct 2, 2023

View reviewed changes

pranjalssh approved these changes Oct 5, 2023

View reviewed changes

feilong-liu force-pushed the scaledwriter_hbo branch 2 times, most recently from dc63547 to af6430c Compare October 6, 2023 18:47

feilong-liu force-pushed the scaledwriter_hbo branch from af6430c to 9e88b78 Compare October 10, 2023 04:59

mlyublena reviewed Oct 10, 2023

View reviewed changes

feilong-liu force-pushed the scaledwriter_hbo branch from 9e88b78 to 47d69a3 Compare October 10, 2023 18:54

feilong-liu added 2 commits October 11, 2023 12:48

Track number of tasks for scaled writer in HBO

1faef8f

Use HBO to set number of tasks for scaled writers

b3f5944

feilong-liu force-pushed the scaledwriter_hbo branch from 47d69a3 to b3f5944 Compare October 11, 2023 19:48

feilong-liu merged commit 18ac248 into prestodb:master Oct 11, 2023

feilong-liu deleted the scaledwriter_hbo branch October 11, 2023 21:51

feilong-liu mentioned this pull request Oct 11, 2023

Add cost based decision log for scaled writer optimizer #21120

Merged

6 tasks

wanglinsong mentioned this pull request Dec 8, 2023

Add release notes for 0.285 #21500

Closed

26 tasks

Conversation

feilong-liu commented Sep 18, 2023 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Description

Motivation and Context

Impact

Test Plan

Contributor checklist

Release Notes

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

3 participants

feilong-liu commented Sep 18, 2023 •

edited

Loading