[
https://issues.apache.org/jira/browse/HDDS-372?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16808414#comment-16808414
]
Tsz Wo Nicholas Sze commented on HDDS-372:
------------------------------------------
Thanks [~shashikant]. Some quick comments:
- Do not change Checksum to use UnsafeByteOperations since (1) checksum size is
very small compared with the data and (2) checksum is used to detect data
change -- if there is a bug involving UnsafeByteOperations, the checksum may be
able to detect it.
- How about renaming the new conf "ozone.safe.buffer.bytestring.copy" to
"ozone.client.UnsafeByteOperations.enabled"?
> There are three buffer copies in BlockOutputStream
> --------------------------------------------------
>
> Key: HDDS-372
> URL: https://issues.apache.org/jira/browse/HDDS-372
> Project: Hadoop Distributed Data Store
> Issue Type: Improvement
> Components: Ozone Client
> Reporter: Tsz Wo Nicholas Sze
> Assignee: Shashikant Banerjee
> Priority: Major
> Attachments: HDDS-372.001.patch, HDDS-372.002.patch,
> HDDS-372.20180829.patch
>
>
> Currently, there are three buffer copies in ChunkOutputStream
> # from byte[] to ByteBuffer, and
> # from ByteBuffer to ByteString.
> # from ByteString to ByteBuffer for checskum computation
> We should eliminate the ByteBuffer in the middle.
> For zero copy io, we should support WritableByteChannel instead of
> OutputStream. It won't be done in this JIRA.
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
---------------------------------------------------------------------
To unsubscribe, e-mail: hdfs-issues-unsubscr...@hadoop.apache.org
For additional commands, e-mail: hdfs-issues-h...@hadoop.apache.org
