Skip to content

feat: Optimize PK MOR read performance by reduce per-row object allocation and improve merge read performance. - #356

Merged
lxy-9602 merged 5 commits into
alibaba:mainfrom
dalingmeng:pk-mor-read-opt
Jun 13, 2026
Merged

feat: Optimize PK MOR read performance by reduce per-row object allocation and improve merge read performance.#356
lxy-9602 merged 5 commits into
alibaba:mainfrom
dalingmeng:pk-mor-read-opt

Conversation

@dalingmeng

Copy link
Copy Markdown
Contributor

Purpose

Linked issue: close #321

Optimize PK MOR read performance by reducing per-row overhead in the merge-on-read path.

This change improves the key-value data file reader and merge functions to avoid unnecessary
object allocation and repeated checks during row iteration. It also adds a fast path when all rows
in a batch are selected, reducing bitmap lookup overhead in common scan scenarios.

Tests

GenericRowTest.TestResetFields

API and Format

Documentation

Generative AI tooling

Generated-by: Aone Copilot, ide-idealab/claude4.6-opus

@dalingmeng dalingmeng changed the title Optimize PK MOR read performance by reduce per-row object allocation and improve merge read performance. feat: Optimize PK MOR read performance by reduce per-row object allocation and improve merge read performance. Jun 10, 2026

@lxy-9602 lxy-9602 left a comment

Copy link
Copy Markdown
Collaborator

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

+1

Comment thread src/paimon/core/io/key_value_data_file_record_reader.h
@dalingmeng
dalingmeng force-pushed the pk-mor-read-opt branch 2 times, most recently from 8f5f33f to 089c7b1 Compare June 11, 2026 03:21

@lxy-9602 lxy-9602 left a comment

Copy link
Copy Markdown
Collaborator

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

+1

@lxy-9602
lxy-9602 merged commit e547186 into alibaba:main Jun 13, 2026
9 of 10 checks passed
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

[Feature] optimize count(*) for PK table

3 participants