[
https://issues.apache.org/jira/browse/HBASE-12148?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15201088#comment-15201088
]
Lars Hofhansl commented on HBASE-12148:
---------------------------------------
I see. So this is at _read_ time from a StoreFile. Is that the main problem?
Not contention at write time.
Is the problem that readers lock each other out? That we can solve with a
simple ReadWrite lock. (but that still takes a memory fence)
Or we can have two TimeRangeTracker implementations. One that mutable (and
synchronized) for the memstore, and one that is immutable (and not synchronized
at all) for StoreFiles.
> Remove TimeRangeTracker as point of contention when many threads writing a
> Store
> --------------------------------------------------------------------------------
>
> Key: HBASE-12148
> URL: https://issues.apache.org/jira/browse/HBASE-12148
> Project: HBase
> Issue Type: Sub-task
> Components: Performance
> Affects Versions: 2.0.0, 0.99.1
> Reporter: stack
> Assignee: Walter Koetke
> Fix For: 2.0.0, 1.3.0, 0.98.19
>
> Attachments:
> 0001-In-AtomicUtils-change-updateMin-and-updateMax-to-ret.patch,
> 12148.addendum.txt, 12148.txt, 12148.txt, 12148v2.txt, 12148v2.txt,
> HBASE-12148-V3.patch, HBASE-12148.txt, HBASE-12148V2.txt, Screen Shot
> 2014-10-01 at 3.39.46 PM.png, Screen Shot 2014-10-01 at 3.41.07 PM.png,
> TimeRangeTracker.tiff
>
>
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
