-
Notifications
You must be signed in to change notification settings - Fork 1.2k
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
Document parallelism and thread scheduling in the architecture guide #8986
Conversation
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Just some minor copy suggestions
@@ -364,8 +368,26 @@ | |||
//! | |||
//! [`RepartitionExec`]: https://docs.rs/datafusion/latest/datafusion/physical_plan/repartition/struct.RepartitionExec.html | |||
//! [Volcano style]: https://w6113.github.io/files/papers/volcanoparallelism-89.pdf | |||
//! [Morsel-Driven Parallelism]: https://db.in.tum.de/~leis/papers/morsels.pdf | |||
//! [DataFusion paper submitted SIGMOD]: https://github.com/apache/arrow-datafusion/files/13874720/DataFusion_Query_Engine___SIGMOD_2024.pdf |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
//! [DataFusion paper submitted SIGMOD]: https://github.com/apache/arrow-datafusion/files/13874720/DataFusion_Query_Engine___SIGMOD_2024.pdf | |
//! [DataFusion paper submitted to SIGMOD]: https://github.com/apache/arrow-datafusion/files/13874720/DataFusion_Query_Engine___SIGMOD_2024.pdf |
Co-authored-by: Raphael Taylor-Davies <[email protected]>
//! DataFusion incrementally computes output from a [`SendableRecordBatchStream`] | ||
//! with `target_partitions` threads. Parallelism is implementing using multiple | ||
//! [Tokio] [`task`]s, which are executed by threads managed by a tokio Runtime. | ||
//! While tokio is most commonly used |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
I should probably also point out that this phrasing came from the SIGMOD paper and I think at least @yjshen and @ozankabak had a part in authoring it
Which issue does this PR close?
Part of #7013
Rationale for this change
The topic of thread scheduling and why not use Rayon or morsel driven parallelism has come up several times, most recently in the Discord channel: https://discord.com/channels/885562378132000778/1166447479609376850/1198721509842235402
I would like to document what is going on so that we can refer people to the documentation when these questions inevitably arise in the future
What changes are included in this PR?
Are these changes tested?
Existing CI checks
Are there any user-facing changes?
More docs