hbase-issues mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Virag Kothari (JIRA)" <j...@apache.org>
Subject [jira] [Commented] (HBASE-12233) Bring back root table
Date Sat, 11 Oct 2014 20:35:34 GMT

    [ https://issues.apache.org/jira/browse/HBASE-12233?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14168328#comment-14168328
] 

Virag Kothari commented on HBASE-12233:
---------------------------------------

bq.  Right now meta gets 20000x more reads than writes on real production clusters

Write throughput on meta become important during cluster startups, operations like create/enable/disable
and during rack failures. Any assignments involved during such activity usually does minimum
2 writes to meta for a single region as each region goes through at least 2 states.
So even in best case scenarios, there will be 2M writes just for bringing a table with 1M
regions online.
Although, a split meta would help with both reads and writes as discussed over in parent JIRA.

> Bring back root table
> ---------------------
>
>                 Key: HBASE-12233
>                 URL: https://issues.apache.org/jira/browse/HBASE-12233
>             Project: HBase
>          Issue Type: Sub-task
>            Reporter: Virag Kothari
>            Assignee: Virag Kothari
>         Attachments: HBASE-12233.patch
>
>
> First step towards splitting meta.



--
This message was sent by Atlassian JIRA
(v6.3.4#6332)

Mime
View raw message