[
https://issues.apache.org/jira/browse/HBASE-9430?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13773841#comment-13773841
]
Lars Hofhansl commented on HBASE-9430:
--------------------------------------
+1 for 0.94 as well.
This is purely cosmetic, though, isn't? If the estimate is wrong a few bytes
per memstore it won't hurt anybody.
What's bad would be if we had a trend (i.e. we forgot to remove or add from/to
the memstore, but we're not doing that here)
> Memstore heapSize calculation - DEEP_OVERHEAD is incorrect
> ----------------------------------------------------------
>
> Key: HBASE-9430
> URL: https://issues.apache.org/jira/browse/HBASE-9430
> Project: HBase
> Issue Type: Bug
> Reporter: Anoop Sam John
> Assignee: Anoop Sam John
> Attachments: HBASE-9430.patch
>
>
> {code}
> expected += ClassSize.estimateBase(ConcurrentSkipListMap.class, false);
> expected += ClassSize.estimateBase(ConcurrentSkipListMap.class, false);
> expected += ClassSize.estimateBase(CopyOnWriteArraySet.class, false);
> expected += ClassSize.estimateBase(CopyOnWriteArrayList.class, false);
> {code}
> We need to consider the heap requirement for KeyValueSkipListSet.
> From where CopyOnWriteArraySet & CopyOnWriteArrayList is coming into picture?
> I am not able to follow.
> Also we need to consider the heap for TimeRangeTracker
--
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
