Fangchen Li created SPARK-55012:
-----------------------------------
Summary: Optimize ThreadPool usage in Arrow UDF concurrent
execution
Key: SPARK-55012
URL: https://issues.apache.org/jira/browse/SPARK-55012
Project: Spark
Issue Type: Improvement
Components: PySpark
Affects Versions: 4.2.0
Reporter: Fangchen Li
The current implementation of Arrow UDF concurrent execution creates a new
{{ThreadPoolExecutor}} for every batch and submits one task for each row. It
can be optimized using a singleton ThreadPool and chunked execution.
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
