-
Notifications
You must be signed in to change notification settings - Fork 379
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
[VL] Fix and use flattenVector #4783
Conversation
Thanks for opening a pull request! Could you open an issue for this pull request on Github Issues? https://github.com/oap-project/gluten/issues Then could you also rename commit message and pull request title in the following format?
See also: |
/Benchmark Velox |
===== Performance report for TPCH SF2000 with Velox backend, for reference only ====
|
/Benchmark Velox |
08ac6b0
to
3e0502f
Compare
@@ -222,6 +222,7 @@ arrow::Status VeloxShuffleWriter::init() { | |||
|
|||
ARROW_ASSIGN_OR_RAISE( | |||
partitioner_, Partitioner::make(options_.partitioning, numPartitions_, options_.startPartitionId)); | |||
DLOG(INFO) << "Create partitioning type: " << std::to_string(options_.partitioning); |
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.
this looks like a debug log?
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.
Yes. The DLOG
ensures that it only get printed with debug mode. We probably need this log for debugging, because sometimes the shuffle operator can get omitted on Spark UI, such as a single partitioning after limit operator.
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.
👍
===== Performance report for TPCH SF2000 with Velox backend, for reference only ====
|
This patch re-enables the flattern vector optimizations.
The flattenVector optimization firstly landed in #4415
but partially reverted in #4474 due to some bugs on Celeborn code path.
Celeborn integration tests should check the code path already