Multi-threaded JRuby BigNum performance ~20x slower than MRI threaded BigNum
----------------------------------------------------------------------------
Key: JRUBY-5214
URL: http://jira.codehaus.org/browse/JRUBY-5214
Project: JRuby
Issue Type: Bug
Components: Performance
Affects Versions: JRuby 1.5.5
Environment: Windows 7 Ultimate, jruby 1.5.5, Java HotSpot VM 1.6.0_22
vs windows MRI 1.9.2. Pik gem used to switch between the two for testing.
Reporter: Aaron Stewart
Attachments: primal.zip
Multi-threaded Miller-Rabin implementation takes ~38 seconds to complete the
included test file on my machine using MRI 1.9.2 - and a single core. Running
the same test file under jRuby takes > 200 seconds despite utilizing 4 cores.
Thread workload is mostly BigNum math.
--fast flag, and large JVM heap sizes, thread pooling alone or in conjunction
did not substantially alter performance.
--
This message is automatically generated by JIRA.
-
If you think it was sent incorrectly contact one of the administrators:
http://jira.codehaus.org/secure/Administrators.jspa
-
For more information on JIRA, see: http://www.atlassian.com/software/jira
---------------------------------------------------------------------
To unsubscribe from this list, please visit:
http://xircles.codehaus.org/manage_email
