[
https://issues.apache.org/jira/browse/HBASE-8635?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13671603#comment-13671603
]
Anoop Sam John edited comment on HBASE-8635 at 5/31/13 4:08 PM:
----------------------------------------------------------------
10% is reasonable. Only thing is we need to consider this also in the above
scenario. May be we need to reduce the block cache max% to 30? [~stack] what
do you say?
bq.It is the upper limit, not reserved all the time.
Ya block cache also upper limit only. My point was what if ,at some point, all
these caches taking its max possible? Will have only 10% remaining memory for
op of the RS. Will that be enough?
was (Author: anoop.hbase):
10% is reasonable. Only thing is we need to consider this also in the above
scenario. May be we need to reduce the block cache max% to 30? [~stack] what
do you?
bq.It is the upper limit, not reserved all the time.
Ya block cache also upper limit only. My point was what if ,at some point, all
these caches taking its max possible? Will have only 10% remaining memory for
op of the RS. Will that be enough?
> 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
