[
https://issues.apache.org/jira/browse/SPARK-56312?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Yicong Huang updated SPARK-56312:
---------------------------------
Description:
Refactor SQL_COGROUPED_MAP_ARROW_UDF to be self-contained in read_udfs(),
moving logic from CogroupArrowUDFSerializer into worker.py, using
ArrowStreamSerializer as pure I/O.
This follows the same pattern as other eval type refactors in this project
(e.g., SPARK-55390, SPARK-56123, SPARK-56189).
was:
Add ASV micro-benchmarks for SQL_COGROUPED_MAP_ARROW_UDF in bench_eval_type.py.
This follows the established pattern from other eval type benchmarks (e.g.,
SPARK-55947 for grouped map, SPARK-56085 for grouped agg).
Summary: Refactor SQL_COGROUPED_MAP_ARROW_UDF to use
ArrowStreamSerializer (was: Add ASV micro-benchmarks for
SQL_COGROUPED_MAP_ARROW_UDF)
> Refactor SQL_COGROUPED_MAP_ARROW_UDF to use ArrowStreamSerializer
> -----------------------------------------------------------------
>
> Key: SPARK-56312
> URL: https://issues.apache.org/jira/browse/SPARK-56312
> Project: Spark
> Issue Type: Sub-task
> Components: PySpark, SQL
> Affects Versions: 4.2.0
> Reporter: Yicong Huang
> Priority: Major
>
> Refactor SQL_COGROUPED_MAP_ARROW_UDF to be self-contained in read_udfs(),
> moving logic from CogroupArrowUDFSerializer into worker.py, using
> ArrowStreamSerializer as pure I/O.
> This follows the same pattern as other eval type refactors in this project
> (e.g., SPARK-55390, SPARK-56123, SPARK-56189).
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
