Skip to content

Optimize performance of ByteViewGroupValueBuilder on batches with inlined views #16330

@Dandandan

Description

@Dandandan

Is your feature request related to a problem or challenge?

Currently a large part of view performance

do_append_val_inner
do_equal_to_inner

and others

is spent rebuilding the views, checking if they can be inlined, etc..

This is wasteful if there is no data buffer, in which case we can improve the performance.

Describe the solution you'd like

Optimize / specialize the code for only inlined view arrays (i.e. without data buffer).

Describe alternatives you've considered

No response

Additional context

No response

Metadata

Metadata

Assignees

Labels

enhancementNew feature or requestperformanceMake DataFusion faster

Type

No type

Projects

No projects

Milestone

No milestone

Relationships

None yet

Development

No branches or pull requests

Issue actions