[
https://issues.apache.org/jira/browse/HBASE-15543?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15216018#comment-15216018
]
stack commented on HBASE-15543:
-------------------------------
The method creates a variety of Cell types dependent on setup. Is the reported
28% creation of new Cell or new Cell reference when the Cell is being made
against a backing BB that is offheap from BucketCache?
bq. But that could be costlier?
[~ram_krish] Why you think? We have to do one copy at least from old backing
BB to the new buffer. There is a further copy being made just so the Cell can
exist outside of this Scanner?
> HFileScannerImpl.getCell() allocates cell object on every call
> --------------------------------------------------------------
>
> Key: HBASE-15543
> URL: https://issues.apache.org/jira/browse/HBASE-15543
> Project: HBase
> Issue Type: Sub-task
> Reporter: Vladimir Rodionov
> Assignee: Vladimir Rodionov
>
> This accounts for 28% of memory allocation in compaction thread.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
