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.

Sign up for GitHub

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

Jump to bottom

continuous benchmarking #263

Open
2 tasks
yliang412 opened this issue Dec 9, 2024 · 1 comment
Open
2 tasks

continuous benchmarking #263

yliang412 opened this issue Dec 9, 2024 · 1 comment

Comments

@yliang412
Copy link
Member

We need to keep track of optd's performance.

Tasks
@yliang412 yliang412 mentioned this issue Dec 9, 2024
Merged
yliang412 added a commit that referenced this issue Dec 10, 2024
@yliang412
plannertest: criterion-rs based benchmark for both planning and exec…
cca20d4 
…ution (#262)

Part of #263.

## Problem 

There is a lack of performance evaluation in optd. Without proper
benchmarking, we will not know if some changes cause significant
regression in performance. We should keep track of the two aspects of
optimizer performance:

- plan generation time 
- actual execution time

## Summary of changes

- Add a [Criterion-rs](https://bheisler.github.io/criterion.rs/book/)
based benchmark to measure optd performance.
- Use the `sqlplannertest` yaml-based setup, bench if there is a `bench`
task.
- Bump `sqlplannertest` to version 0.4.
- better individual test matching added in
risinglightdb/sqlplannertest-rs#9

Planning:
  - Input: SQL statement (aka `datafusion_sql::parser::Statement`).
  - Measures the time it takes to generate a datafusion physical plan.

Execution:
- Input: datafusion physical plan outputted from optd (aka `Arc<dyn
ExecutionPlan>`).
- Measures the time it takes to execute the entire query.

## Next Steps
https://github.com/benchmark-action/github-action-benchmark can directly
take `cargo bench` results as input. So it will be easy to set up CI.

---------

Signed-off-by: Yuchen Liang <[email protected]>
@epompeii
Copy link

@yliang412 if you are looking to use Criterion + github-action-benchmark, you will need to run in compat mode: benchmark-action/github-action-benchmark#8
Depending on your use case this may not be desirable.

I'm more than happy to help answer any questions to get you all set up with Bencher, which has full Criterion support.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Assignees
No one assigned
Labels
None yet
Projects
None yet
Milestone
No milestone
Development

No branches or pull requests
2 participants
@epompeii @yliang412