[
https://issues.apache.org/jira/browse/HDDS-2713?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Attila Doroszlai resolved HDDS-2713.
------------------------------------
Fix Version/s: 0.6.0
Resolution: Implemented
[~arp], yes, this work is done. Only optional enhancement is left, which can
be implemented later.
> Decouple client write size and datanode chunk size
> --------------------------------------------------
>
> Key: HDDS-2713
> URL: https://issues.apache.org/jira/browse/HDDS-2713
> Project: Hadoop Distributed Data Store
> Issue Type: Improvement
> Components: Ozone Datanode, SCM Client
> Reporter: Attila Doroszlai
> Assignee: Attila Doroszlai
> Priority: Major
> Labels: TriagePending
> Fix For: 0.6.0
>
>
> Currently Datanode creates chunk files as they are received from HDDS client.
> This creates a conflict between requirements: client would like to use less
> memory for buffering, but datanode needs to avoid small files.
> The goal of this task is to decouple client and server write sizes to allow
> the client to send data in smaller increments without affecting datanode
> storage.
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
