[
https://issues.apache.org/jira/browse/DERBY-2911?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Knut Anders Hatlen updated DERBY-2911:
--------------------------------------
Attachment: perftest.pdf
perftest2.diff
I finally got around to look at the results from the tests I had running. Sorry
for the delay.
Attached is a diff (perftest2.diff) with an updated version of the performance
tests that I used (the load that were supposed to resemble the d2911perf class
attached to this issue, was changed so that it matched d2911perf more closely,
and this change also simplified the code a bit), and a pdf (perftest.pdf) which
summarizes the findings. Nothing new there, actually, the results were pretty
consistent with the results that have been posted here earlier. The new buffer
manager performs significantly better than the old one when there's no other
contention point (like the log device or the root node in the B-tree) and there
are multiple CPUs/cores, and a little better than or equal to the old buffer
manager in cases where there is another contention point.
> Implement a buffer manager using java.util.concurrent classes
> -------------------------------------------------------------
>
> Key: DERBY-2911
> URL: https://issues.apache.org/jira/browse/DERBY-2911
> Project: Derby
> Issue Type: Improvement
> Components: Performance, Services
> Affects Versions: 10.4.0.0
> Reporter: Knut Anders Hatlen
> Assignee: Knut Anders Hatlen
> Priority: Minor
> Attachments: cleaner.diff, cleaner.tar, d2911-1.diff, d2911-1.stat,
> d2911-10.diff, d2911-10.stat, d2911-11.diff, d2911-2.diff, d2911-3.diff,
> d2911-4.diff, d2911-5.diff, d2911-6.diff, d2911-6.stat, d2911-7.diff,
> d2911-7a.diff, d2911-9.diff, d2911-9.stat, d2911-entry-javadoc.diff,
> d2911-unused.diff, d2911-unused.stat, d2911perf.java, derby-2911-8.diff,
> derby-2911-8.stat, perftest.diff, perftest.pdf, perftest.stat,
> perftest2.diff, perftest6.pdf, poisson_patch8.tar
>
>
> There are indications that the buffer manager is a bottleneck for some types
> of multi-user load. For instance, Anders Morken wrote this in a comment on
> DERBY-1704: "With a separate table and index for each thread (to remove latch
> contention and lock waits from the equation) we (...) found that
> org.apache.derby.impl.services.cache.Clock.find()/release() caused about 5
> times more contention than the synchronization in LockSet.lockObject() and
> LockSet.unlock(). That might be an indicator of where to apply the next push".
> It would be interesting to see the scalability and performance of a buffer
> manager which exploits the concurrency utilities added in Java SE 5.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
