[ 
https://issues.apache.org/jira/browse/HBASE-12934?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14294611#comment-14294611
 ] 

Andrew Purtell edited comment on HBASE-12934 at 1/28/15 2:32 AM:
-----------------------------------------------------------------

I didn't misread the patch. :-) I don't think we want to put all of the data 
for the whole CF on one tier, we should be deciding based on access statistics 
whether to migrate HFiles from one tier to the other, not just assuming a flush 
or freshly compacted file will benefit. Any given file could end up just being 
cold data. I would agree that a coarse grain "policy" of flushing and 
compacting every file of a CF would be a first step, and it might be useful if 
a user can judge it properly for their use case, but this would be another 
thing the user has to specify correctly in the schema without possibility of 
autotuning/dynamic decision making. So I don't think it's quite the right 
approach.


was (Author: apurtell):
I didn't misread the patch. :-) I don't think we want to put all of the data 
for the whole CF on one tier, we should be deciding based on access statistics 
whether to migrate HFiles from one tier to the other, not just assuming a flush 
or freshly compacted file will benefit. Any given file could end up just being 
cold data. I would agree that a coarse grain "policy" of flushing and 
compacting every file of a CF would be a first step, and it might be useful if 
a user can judge it properly for their use case, but this would be another 
thing the user has to specify correctly in the schema without possibility of 
autotuning/dynamic decision making.

> Utilize Flash storage for flushing
> ----------------------------------
>
>                 Key: HBASE-12934
>                 URL: https://issues.apache.org/jira/browse/HBASE-12934
>             Project: HBase
>          Issue Type: Sub-task
>            Reporter: Ted Yu
>            Assignee: Ted Yu
>         Attachments: 12934-001.txt
>
>
> Store flushing should be able to make use of hdfs storage policy.
> One option is to allow setting storage policy for the directory path of the 
> specified column family.



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

Reply via email to