[
https://issues.apache.org/jira/browse/HBASE-10015?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13829686#comment-13829686
]
Lars Hofhansl commented on HBASE-10015:
---------------------------------------
I think it would be hard to
# push reader switching to the scanning thread
# guarantee that the readers will eventually be switched to the compaction can
proceed and finished
What if a scanner isn't closed? Or nobody ever calls next() on it? Then we have
to wait for the lease to expire before the compaction can finish.
next()/seek()/reseek() are not as critical as they are not called remotely as
often as peek. peek() is also a very small method and (according to our perf
expert here) small synchronized methods have the greatest potential to throw
the branch prediction off.
> Major performance improvement: Avoid synchronization in StoreScanner
> --------------------------------------------------------------------
>
> Key: HBASE-10015
> URL: https://issues.apache.org/jira/browse/HBASE-10015
> Project: HBase
> Issue Type: Bug
> Reporter: Lars Hofhansl
> Assignee: Lars Hofhansl
> Fix For: 0.98.0, 0.96.1, 0.94.15
>
> Attachments: 10015-0.94-v2.txt, 10015-0.94-v3.txt,
> 10015-0.94-withtest.txt, 10015-0.94.txt, 10015-trunk-v2.txt,
> 10015-trunk-v3.txt, 10015-trunk.txt, TestLoad.java
>
>
> Did some more profiling (this time with a sampling profiler) and
> StoreScanner.peek() showed up a lot in the samples. At first that was
> surprising, but peek is synchronized, so it seems a lot of the sync'ing cost
> is eaten there.
> It seems the only reason we have to synchronize all these methods is because
> a concurrent flush or compaction can change the scanner stack, other than
> that only a single thread should access a StoreScanner at any given time.
> So replaced updateReaders() with some code that just indicates to the scanner
> that the readers should be updated and then make it the using thread's
> responsibility to do the work.
> The perf improvement from this is staggering. I am seeing somewhere around 3x
> scan performance improvement across all scenarios.
> Now, the hard part is to reason about whether this is 100% correct. I ran
> TestAtomicOperation and TestAcidGuarantees a few times in a loop, all still
> pass.
> Will attach a sample patch.
--
This message was sent by Atlassian JIRA
(v6.1#6144)
