axis-java-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Glen Daniels <g...@thoughtcraft.com>
Subject Re: [Axis2] AXIOM vs. JDOM trivial perf test
Date Thu, 30 Mar 2006 21:36:18 GMT
Hi dims:

Thanks for looking into this.  It appears the results are pretty 
dependent on the type of XML file (and clearly also JVM).  My test file 
was element-heavy with not much content, where the test results are very 
content-skewed.  Try your test with a file that's mostly element structure:

<foo:test xmlns:foo="http://namespace">
   <elem attr="something">
     <foo:subElem>some text</foo:subElem>
   </elem>
   ...repeat a bajillion times...
</foo:test>

With this kind of thing I'm seeing JDOM do better and better with 
respect to AXIOM as the # of elements grows.  I'd be curious to see if 
JDK affects that.  I'll post the file and test in the JIRA Dennis opened.

I did try the tests with the IBM-Axis result file, and I see a MUCH less 
pronounced difference (77ms for OM vs 70ms for JDOM).

--Glen

Mime
View raw message