[
https://issues.apache.org/jira/browse/HDDS-4454?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17233176#comment-17233176
]
runzhiwang commented on HDDS-4454:
----------------------------------
[~shashikant] Thanks for the great work. I think this will improve the write
efficiency significantly. Besides, ratis streaming is mainly used for Big
file. If the file is small, such as smaller than 4MB (the size of chunk). We
should use the origin API, because ratis streaming need more rpc call. See here:
!screenshot-1.png!
> Ozone Streaming Write Pipeline
> ------------------------------
>
> Key: HDDS-4454
> URL: https://issues.apache.org/jira/browse/HDDS-4454
> Project: Hadoop Distributed Data Store
> Issue Type: New Feature
> Components: Ozone Client, Ozone Datanode
> Reporter: Shashikant Banerjee
> Assignee: Shashikant Banerjee
> Priority: Major
> Attachments: Ozone Write Streaming.pdf, screenshot-1.png
>
>
> https://issues.apache.org/jira/browse/RATIS-979 implements streaming
> semantics in Ratis.
> Currently, ozone write pattern is bursty and involves multiple buffers copies
> as well multiple ratis log syncs in a block write. The idea of the Jira is to
> use a zero buffer copy based ratis streaming in ozone write path for better
> performance and resource utilisation.
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
