harmony-commits mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Li-Gang Wang (JIRA)" <j...@apache.org>
Subject [jira] Commented: (HARMONY-4891) [drlvm][gc] heap exhaustion performance is five times slower than RI's one
Date Tue, 16 Oct 2007 14:16:50 GMT

    [ https://issues.apache.org/jira/browse/HARMONY-4891?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel#action_12535253
] 

Li-Gang Wang commented on HARMONY-4891:
---------------------------------------

First of all, I think it is more meaningful to experiment with the same max heap size with
two VMs.

In my test, the execution times of GCTest with -Xmx64m setting of the two VMs are very close:
DRLVM(46220ms), RI(46588ms).
With -Xmx256m setting we can see the perf of DRLVM falls into the 70% scale (actually 87.6%).

The rest of the problem is why we set 256m as the default heap size rather than 64m as RI
does. I think either setting has its pros and cons. For instance, with default heap size,
DRLVM can finish running 4 warehouses when running SPECjbb2005, while RI cannot even finish
loading the 3rd warehouse. So I think we need a full-scale consideration on this problem.
While it might be another issue.

I also ran the newly attached simple test with -Xmx64m on both my desktop PC and notebook,
and the test ran well. Alexei, could you please list your environment parameters and the command
line here? Thanks.

> [drlvm][gc] heap exhaustion performance is five times slower than RI's one
> --------------------------------------------------------------------------
>
>                 Key: HARMONY-4891
>                 URL: https://issues.apache.org/jira/browse/HARMONY-4891
>             Project: Harmony
>          Issue Type: Bug
>          Components: DRLVM
>         Environment: Windows, ia32
>            Reporter: Alexei Fedotov
>            Assignee: Xiao-Feng Li
>         Attachments: AbstractTest.java, FreeMem.java, GcTest.java, p-unit-0.12.jar, p-unit-extension-0.12.jar
>
>
> $ time win_ia32_msvc_release/deploy/jdk/jre/bin/java -cp bin/classes org.apache.harmony.test.stress.gc.share.GcTest
> [concurrent] Starting org.apache.harmony.test.stress.gc.share.GcTest
> [debug] total = 268435456, used = 2165188, ratio = 13, sizes = 1 - 2097152
> test() - [123709.342995ms,97.95109033584595%]
> total: 1, failures:0 (GREEN) - 123770.125835ms
> real    2m9.328s
> user    0m0.015s
> sys     0m0.000s
> $ time java -cp bin/classes org.apache.harmony.test.stress.gc.share.GcTest
> [concurrent] Starting org.apache.harmony.test.stress.gc.share.GcTest
> [debug] total = 66650112, used = 142120, ratio = 13, sizes = 1 - 2097152
> test() - [47643.880758ms,92.48719041912487%]
> total: 1, failures:0 (GREEN) - 47670.789275ms
> real    0m47.953s
> user    0m0.015s
> sys     0m0.000s
> The test itself runs exactly 30sec, and most of the time is taken by single-threaded
heap exhaustions to estimate the maximum size available for allocation.
> 1. It would be great to get within 70% of RI performance.
> 2. It would be great if one tunes debug build performance which is currently x40 times
slower. The debug build is used for regular stress test runs and this would help to save some
time.

-- 
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.


Mime
View raw message