[
https://issues.apache.org/jira/browse/SPARK-25115?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16580106#comment-16580106
]
Apache Spark commented on SPARK-25115:
--------------------------------------
User 'normanmaurer' has created a pull request for this issue:
https://github.com/apache/spark/pull/22105
> Eliminate extra memory copy done when a ByteBuf is used that is backed by
> > 1 ByteBuffer.
> ---------------------------------------------------------------------------------------------
>
> Key: SPARK-25115
> URL: https://issues.apache.org/jira/browse/SPARK-25115
> Project: Spark
> Issue Type: Improvement
> Components: Input/Output
> Affects Versions: 2.3.1
> Reporter: Norman Maurer
> Priority: Major
>
> Calling ByteBuf.nioBuffer(...) can be costly when the ByteBuf is backed by
> more then 1 ByteBuf. In this case it makes more sense to call
> ByteBuf.nioBuffers(...) and iterate over the returned ByteBuffer[].
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
---------------------------------------------------------------------
To unsubscribe, e-mail: issues-unsubscr...@spark.apache.org
For additional commands, e-mail: issues-h...@spark.apache.org
