[WIP] Speed up deserialization of object refs #17882
Closed
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Outdated suggestions cannot be applied.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.
Why are these changes needed?
In #17803 it is reported that calling
ray.get()
on object refs contained within an object can be expectedly slow.Cache Python call site when creating object refs contained in the same deserializing object. Getting Python call site is the most expensive operation when creating object refs. Also move some deserialization logic into
cython
.Add a microbenchmark for
ray.get()
on objects containing 10k object refs. Onm5.8xlarge
,Before:
single client get object containing 10k refs per second 6.68 +- 0.01
After:
single client get object containing 10k refs per second 17.52 +- 0.8
Related issue number
#17803
Checks
scripts/format.sh
to lint the changes in this PR.