hbase-issues mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Nicolas Liochon (JIRA)" <j...@apache.org>
Subject [jira] [Commented] (HBASE-10355) Failover RPC's from client using region replicas
Date Wed, 19 Feb 2014 19:02:21 GMT

    [ https://issues.apache.org/jira/browse/HBASE-10355?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13905878#comment-13905878
] 

Nicolas Liochon commented on HBASE-10355:
-----------------------------------------

bq. We should fix either that or this
I think we need micro seconds, at least for testing: a synchronous get takes ~0.2ms on my
1Gb network, so if we want to play with the secondary calls probability with probability different
than 0 & 100%, we need it. As well, for end user, it makes send to use values like 50m
microseconds: compared to 1ms, it divides the latency by 2. And the perf will be much better
on a modern hardware actually.

bq. It works with HBASE-10472, right?
Yes, plus HBASE-10525

bq. I guess CompletionService is also available for jdk6
Actually yes. I was wrong. But the v2 simplifies a little the code compared to the default
CompletionService, so I prefer to keep it actually.

> Failover RPC's from client using region replicas
> ------------------------------------------------
>
>                 Key: HBASE-10355
>                 URL: https://issues.apache.org/jira/browse/HBASE-10355
>             Project: HBase
>          Issue Type: Sub-task
>          Components: Client
>            Reporter: Enis Soztutar
>            Assignee: Nicolas Liochon
>             Fix For: 0.99.0
>
>         Attachments: 10355.v1.patch, 10355.v2.patch
>
>




--
This message was sent by Atlassian JIRA
(v6.1.5#6160)

Mime
View raw message