[FEAT] Ray shuffle experiment #2883
Closed
Conversation
This file contains 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
CodSpeed Performance ReportMerging #2883 will not alter performanceComparing Summary
|
added 9 commits
September 28, 2024 18:31
…a if this works though
jaychia
force-pushed
the
jay/shuffle-experiment
branch
from
6b5a053 to
cef8843
Compare
Codecov ReportAttention: Patch coverage is
Additional details and impacted files@@ Coverage Diff @@
## main #2883 +/- ##
==========================================
- Coverage 78.71% 78.38% -0.34%
==========================================
Files 597 599 +2
Lines 69735 70131 +396
==========================================
+ Hits 54895 54969 +74
- Misses 14840 15162 +322
Flags with carried forward coverage won't be shown. Click here to find out more.
|
jaychia
added a commit
that referenced
this pull request
Oct 28, 2024
…an variant (#3083) # Summary Refactors our `PhysicalPlan` by adding a new `PhysicalPlan::ShuffleExchange` variant. ## Why? **This PR makes it easier to add new shuffle implementations, by simply extending the `PhysicalPlan::ShuffleExchange` enum** 1. To users, the physical plan looks a lot more legible (even I don't really know what a "Flatten" is...) 2. Easier to add new types of shuffles... Also, there's a nice place to add a layer of heuristics to figure out when to use the new shuffles (the new `ShuffleExchangeBuilder` builder can let us direct clients to use more complex shuffles by consulting environment variables, available cluster resources, number of partitions etc) ## Elaboration Our previous PhysicalPlan was a bit of a leaky abstraction, expressing 2 types of shuffles by invoking an unnecessarily low-level chain of operations to users of the abstraction, as well as users of Daft: 1. `FanoutHash/Range/Random -> Flatten -> ReduceMerge`: this pattern was used to express a "NaiveFullyMaterializingMapReduce" shuffle (materialize all Map tasks, perform a "flatten" which makes no sense to users, then reduce and merge) 3. `Split/Coalesce -> Flatten`: this pattern was used to express a "SplitOrCoalesceToTargetNum" shuffle This leaky abstraction makes it difficult to add new types of shuffles (e.g. the push-based shuffle implemented in #2883) as it involves adding new PhysicalPlan variant(s). Additionally, all of these shuffles share similar characteristics during plan optimization, and the actual implementation of "how" to execute these shuffles should be highly dependent on factors such as available cluster resources, expected complexity of the shuffle and more. --------- Co-authored-by: Jay Chia <[email protected]@users.noreply.github.com>
|
@jaychia is PR still relevant after merging in your other one? |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
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.
No description provided.