Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

coldata,sql: remove some todos #53588

Merged
merged 1 commit into from
Aug 31, 2020
Merged

Conversation

yuzefovich
Copy link
Member

This commit removes several TODOs that I have prototyped addressing and
decided to abandon the prototypes, namely:

  • checking whether coldata.BatchSize() atomic has influence on
    performance (the benchmarks and TPCH queries showed that the impact is
    negligible if any)
  • tuning default batch size (I did that a while ago, and the best
    batch size according tpchvec/bench was 1280, barely better than current
    1024 which is a lot nicer number)
  • pooling allocations of execFactory objects (this showed some
    improvement on one workload and a regression on another).

Release justification: non-production code changes.

Release note: None

@cockroach-teamcity
Copy link
Member

This change is Reviewable

@yuzefovich
Copy link
Member Author

Benchmarks of constant non-atomic batch size are here.

Benchmarks of pooling exec factories are here.

Copy link
Contributor

@asubiotto asubiotto left a comment

Choose a reason for hiding this comment

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

:lgtm:

Reviewed 2 of 2 files at r1.
Reviewable status: :shipit: complete! 1 of 0 LGTMs obtained (waiting on @yuzefovich)


pkg/col/coldata/batch.go, line 75 at r1 (raw file):

var _ Batch = &MemBatch{}

// TODO(jordan): tune.

Add a comment about your benchmarks for future readers

This commit removes several TODOs that I have prototyped addressing and
decided to abandon the prototypes, namely:
- checking whether `coldata.BatchSize()` atomic has influence on
performance (the benchmarks and TPCH queries showed that the impact is
negligible if any)
- tuning default batch size (I did that a while ago, and the best
batch size according tpchvec/bench was 1280, barely better than current
1024 which is a lot nicer number)
- pooling allocations of `execFactory` objects (this showed some
improvement on one workload and a regression on another).

Release justification: non-production code changes.

Release note: None
Copy link
Member Author

@yuzefovich yuzefovich left a comment

Choose a reason for hiding this comment

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

TFTR!

bors r+

Reviewable status: :shipit: complete! 0 of 0 LGTMs obtained (and 1 stale) (waiting on @asubiotto)


pkg/col/coldata/batch.go, line 75 at r1 (raw file):

Previously, asubiotto (Alfonso Subiotto Marqués) wrote…

Add a comment about your benchmarks for future readers

Done.

@craig
Copy link
Contributor

craig bot commented Aug 31, 2020

Build failed (retrying...):

@craig
Copy link
Contributor

craig bot commented Aug 31, 2020

Build succeeded:

@craig craig bot merged commit 42e73f7 into cockroachdb:master Aug 31, 2020
@yuzefovich yuzefovich deleted the remove-todos branch August 31, 2020 20:02
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.

3 participants