weDataSphere created SPARK-27615:
------------------------------------
Summary: Merge small files in the read stage
Key: SPARK-27615
URL: https://issues.apache.org/jira/browse/SPARK-27615
Project: Spark
Issue Type: New Feature
Components: Spark Core
Affects Versions: 2.4.0
Reporter: weDataSphere
When Spark reads file paths to form RDDs, the more files there are, the more
partitions there will be. RDD partition can improve computing speed to a
certain extent, but too much parity will lead to too much Task scheduling speed
reduction, and generate more files.
We can reduce the number of partitions by judging whether a file can be merged
during Spark reading and, if so, merged reading.
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
