[
https://issues.apache.org/jira/browse/HBASE-5898?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13529606#comment-13529606
]
Lars Hofhansl commented on HBASE-5898:
--------------------------------------
Possibly. This will very much look like a deadlock. Scanners will time out, and
it will look like the system has ground to a halt.
This will be exasperated by large store files. Imagine a 20GB store file and
only one reader at a time allowed on it.
I think we can have a adhoc fix where attempt seek + read and if the istream is
locked we'll switch to pread. That way we get seek + read when possible and
pread when necessary.
There also *might* be a mysterious deadlock, but until I see a jstack or am
shown code that causes it I'll be skeptical.
> Consider double-checked locking for block cache lock
> ----------------------------------------------------
>
> Key: HBASE-5898
> URL: https://issues.apache.org/jira/browse/HBASE-5898
> Project: HBase
> Issue Type: Improvement
> Components: Performance
> Affects Versions: 0.94.1
> Reporter: Todd Lipcon
> Assignee: Todd Lipcon
> Priority: Critical
> Fix For: 0.94.3, 0.96.0
>
> Attachments: 5898-0.94.txt, 5898-TestBlocksRead.txt, 5898-v2.txt,
> 5898-v3.txt, 5898-v4.txt, 5898-v4.txt, HBASE-5898-0.patch,
> HBASE-5898-1.patch, HBASE-5898-1.patch, hbase-5898.txt
>
>
> Running a workload with a high query rate against a dataset that fits in
> cache, I saw a lot of CPU being used in IdLock.getLockEntry, being called by
> HFileReaderV2.readBlock. Even though it was all cache hits, it was wasting a
> lot of CPU doing lock management here. I wrote a quick patch to switch to a
> double-checked locking and it improved throughput substantially for this
> workload.
--
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
