[
https://issues.apache.org/jira/browse/HDFS-12976?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16310398#comment-16310398
]
Konstantin Shvachko commented on HDFS-12976:
--------------------------------------------
Thanks for the link [~elgoiri]. Yes something similar to what you did for RBF,
but we will need more sophisticated logic than just random. The trade-offs I
see:
* Switch over from active to standby could be expensive since the client should
wait while SBN catches up with its lastSeenId, so {{StandbyReadProxyProvider}}
should be sticky once it starts reading from SBN.
* But we also want load balancing so that the overall load is distributed
evenly between nodes.
> Introduce StandbyReadProxyProvider
> ----------------------------------
>
> Key: HDFS-12976
> URL: https://issues.apache.org/jira/browse/HDFS-12976
> Project: Hadoop HDFS
> Issue Type: Sub-task
> Components: hdfs-client
> Reporter: Konstantin Shvachko
>
> {{StandbyReadProxyProvider}} should implement {{FailoverProxyProvider}}
> interface and be able to submit read requests to ANN and SBN(s).
--
This message was sent by Atlassian JIRA
(v6.4.14#64029)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
