Add JMH benchmark to measure multi-producer batch publication performance. #483
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Outdated suggestions cannot be applied.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.
Batch publication has much more complicated logic in the
MultiProducerSequencer
compared to theSingleProducerSequencer
where the cursor is simply updated with the high sequence value.Therefore, there is also room for optimizations! :-)
Specifically, I am working on an idea to use just a single bit per slot in the ring buffer to represent "publication status" rather than an
int
and it should have the same performance for single event publication but better performance for the batch publication case.I've already seen great results with this approach in the Rust port I've written (https://github.com/nicholassm/disruptor-rs) and I would like to explore the same idea in the "original" Disruptor.
But first it would be nice with a JMH benchmark to actually measure things.
That's what this small Pull-Request adds. On my laptop, batch publication has 3 times higher throughput compared to single event publication. Can we do better?