starrocks icon indicating copy to clipboard operation
starrocks copied to clipboard

[Enhancement] Scale writers for connector sink when writing static partition tables

Open mxdzs0612 opened this issue 1 year ago • 5 comments

Why I'm doing:

What I'm doing:

By default, writers which quantity is equal to sink_dop will be created at the same time, leading to small files when there is little data. This PR scales up writers one by one for static partition tables, which can cope with this problem.

Before: image

After: image

What type of PR is this:

  • [ ] BugFix
  • [ ] Feature
  • [x] Enhancement
  • [ ] Refactor
  • [ ] UT
  • [ ] Doc
  • [ ] Tool

Does this PR entail a change in behavior?

  • [ ] Yes, this PR will result in a change in behavior.
  • [x] No, this PR will not result in a change in behavior.

If yes, please specify the type of change:

  • [ ] Interface/UI changes: syntax, type conversion, expression evaluation, display information
  • [ ] Parameter changes: default values, similar parameters but with different default values
  • [ ] Policy changes: use new policy to replace old one, functionality automatically enabled
  • [ ] Feature removed
  • [ ] Miscellaneous: upgrade & downgrade compatibility, etc.

Checklist:

  • [ ] I have added test cases for my bug fix or my new feature
  • [ ] This pr needs user documentation (for new or modified features or behaviors)
    • [ ] I have added documentation for my new feature or new function
  • [ ] This is a backport pr

Bugfix cherry-pick branch check:

  • [x] I have checked the version labels which the pr will be auto-backported to the target branch
    • [x] 3.2
    • [ ] 3.1
    • [ ] 3.0
    • [ ] 2.5

mxdzs0612 avatar Feb 01 '24 07:02 mxdzs0612

@mxdzs0612 Nice work! Do you have any performance number before and after applying this PR?

DorianZheng avatar Feb 26 '24 09:02 DorianZheng

@mxdzs0612 Nice work! Do you have any performance number before and after applying this PR?

image

Just a brief test in SF10. there will be less differences in larger scales.

mxdzs0612 avatar Feb 27 '24 02:02 mxdzs0612

[FE Incremental Coverage Report]

:white_check_mark: pass : 2 / 2 (100.00%)

file detail

path covered_line new_line coverage not_covered_line_detail
:large_blue_circle: com/starrocks/qe/SessionVariable.java 2 2 100.00% []

github-actions[bot] avatar Mar 26 '24 04:03 github-actions[bot]

[BE Incremental Coverage Report]

:white_check_mark: pass : 23 / 26 (88.46%)

file detail

path covered_line new_line coverage not_covered_line_detail
:large_blue_circle: be/src/exec/pipeline/exchange/local_exchange.cpp 10 13 76.92% [363, 364, 368]
:large_blue_circle: be/src/runtime/iceberg_table_sink.cpp 1 1 100.00% []
:large_blue_circle: be/src/exec/pipeline/pipeline_builder.cpp 6 6 100.00% []
:large_blue_circle: be/src/exec/chunk_buffer_memory_manager.h 1 1 100.00% []
:large_blue_circle: be/src/runtime/hive_table_sink.cpp 1 1 100.00% []
:large_blue_circle: be/src/exec/pipeline/exchange/local_exchange.h 3 3 100.00% []
:large_blue_circle: be/src/runtime/table_function_table_sink.cpp 1 1 100.00% []

github-actions[bot] avatar Mar 26 '24 04:03 github-actions[bot]

@satanson @dirtysalt PTAL, thx~

mxdzs0612 avatar Mar 26 '24 06:03 mxdzs0612