Skip to content

[Core] Optimize Parquet Reading by Enhancing Predicate Pushdown to the Page Level#4587

Closed
Aiden-Dong wants to merge 1 commit intoapache:masterfrom
Aiden-Dong:master
Closed

[Core] Optimize Parquet Reading by Enhancing Predicate Pushdown to the Page Level#4587
Aiden-Dong wants to merge 1 commit intoapache:masterfrom
Aiden-Dong:master

Conversation

@Aiden-Dong
Copy link
Copy Markdown
Contributor

Purpose

Linked issue: #4586

Tests

API and Format

Documentation

@JingsongLi
Copy link
Copy Markdown
Contributor

This modification is not that simple, it requires us to modify the column reader.

@Aiden-Dong Aiden-Dong changed the title Optimize Parquet Reading by Enhancing Predicate Pushdown to the Page Level [Core] Optimize Parquet Reading by Enhancing Predicate Pushdown to the Page Level Nov 25, 2024
@Aiden-Dong
Copy link
Copy Markdown
Contributor Author

This modification is not that simple, it requires us to modify the column reader.

I found that it works in my local tests, and the column pages obtained based on FilterRowGroup are filtered. Is there any error I might have missed?

@JingsongLi
Copy link
Copy Markdown
Contributor

#3610

@JingsongLi
Copy link
Copy Markdown
Contributor

See implementation in Spark.

@JingsongLi
Copy link
Copy Markdown
Contributor

Close this now, feel free to re-open if you have more questions.

@JingsongLi JingsongLi closed this Nov 26, 2024
@Aiden-Dong
Copy link
Copy Markdown
Contributor Author

See implementation in Spark.

ok tks, I'll go check how parquet is implemented when reading in Spark.

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.

2 participants