[
https://issues.apache.org/jira/browse/HBASE-2926?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12900592#action_12900592
]
Jeff Hammerbacher commented on HBASE-2926:
------------------------------------------
Seems like a reasonable thing to do. I hear a lot of concernts about HDFS from
HBase folks, and it would be great to gather up the concerns in a single place
so we can start tackling them one by one!
> Document about the characteristic of HBase use HDFS
> ---------------------------------------------------
>
> Key: HBASE-2926
> URL: https://issues.apache.org/jira/browse/HBASE-2926
> Project: HBase
> Issue Type: Task
> Reporter: jinglong.liujl
> Original Estimate: 120h
> Remaining Estimate: 120h
>
> As our know, HDFS is design for mapreduce (streaming read/write) , and
> for HBase something can not fit for currently requirement. As Lipcon says,
> random read, sync and client fault tolerant is the most important things we
> should take care of. The document will make some discusses from these three
> points.
>
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
