[
https://issues.apache.org/jira/browse/HBASE-8635?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13672280#comment-13672280
]
Jimmy Xiang commented on HBASE-8635:
------------------------------------
bq. worries me is all those users that already use 80% of their heap with BC
and MemStores that won't be able to restart HBase once they upgrade because
they'd now be at 90% with pre-fetching.
I will make sure prefetching is in checkForClusterFreeMemoryLimit. Fortunately,
this is for 0.96. I agree it is a concern, but it will be a little less of
concern.
I think that the fixed number is not very nice, either 256MB or 128MB. I think
it is good to scale with the amount of memory given to HBase. If this is not
wanted, the percentage can be adjusted. BC is very helpful if data is accessed
multiple times. Prefetching, however, should help all the time. As Anoop
suggested, we can lower the default BC to 30%.
> Define prefetcher.resultsize.max as percentage
> ----------------------------------------------
>
> Key: HBASE-8635
> URL: https://issues.apache.org/jira/browse/HBASE-8635
> Project: HBase
> Issue Type: Improvement
> Reporter: Ted Yu
> Assignee: Jimmy Xiang
> Priority: Minor
> Attachments: trunk-8635.patch
>
>
> Currently "hbase.hregionserver.prefetcher.resultsize.max" defines global
> limit for prefetching.
> The default value is 256MB.
> It would be more flexible to define this measure as a percentage of the heap.
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators
For more information on JIRA, see: http://www.atlassian.com/software/jira
