Add ability to set observeOn buffer size per stage#800
Open
timmartin-stripe wants to merge 2 commits intoNetflix:masterfrom
Open
Add ability to set observeOn buffer size per stage#800timmartin-stripe wants to merge 2 commits intoNetflix:masterfrom
timmartin-stripe wants to merge 2 commits intoNetflix:masterfrom
Conversation
We've found that these buffers are very important tuning parameters. Additionally, they are very stage specific. Sometimes, you need want small buffers and other times large buffers depending on the performance characteristics of the stage. Setting them all via the default RX configuration is not a strong strategy. Especially because that needs to be set when mantis-agent spins up; not necessarily when the job spins up. This commit is mostly plumbing to get the buffer size to the stage executors.
| .map(observable -> c | ||
| .call(context, observable | ||
| .observeOn(Schedulers.computation()) | ||
| .observeOn(Schedulers.computation(), bufferSize) |
Contributor
Author
There was a problem hiding this comment.
This is the key update. The default will be the same rx.internal.util.RxRingBuffer.SIZE
timmartin-stripe
requested review from
Andyz26,
calvin681,
dtrager02,
fdc-ntflx,
hellolittlej and
james-lubin
as code owners
timmartin-stripe
had a problem deploying
to
Integrate Pull Request
GitHub Actions
Failure
| // copy reference, then apply request specific filters, sampling | ||
| Observable<T> requestObservable = observableToServe; | ||
|
|
||
| // decouple the observable on a separate thread and add backpressure handling |
Contributor
Author
There was a problem hiding this comment.
Never used. Figured it was good to clean up
Andyz26
approved these changes
Oct 24, 2025
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Context
We've found that these buffers are very important tuning parameters. Additionally, they are very stage specific. Sometimes, you need want small buffers and other times large buffers depending on the performance characteristics of the stage. Setting them all via the default RX configuration is not a strong strategy. Especially because that needs to be set when mantis-agent spins up; not necessarily when the job spins up.
This commit is mostly plumbing to get the buffer size to the stage executors.
Checklist
./gradlew buildcompiles code correctly./gradlew testpasses all tests