[jdom-interest] Performance tests

philip.nelson at omniresources.com philip.nelson at omniresources.com
Wed May 9 10:48:17 PDT 2001


> - You prime JDOM, but not DOM.

Look a little closer, prime happens for both.  I *clearly* stated that I
made the assumption that using saxbuilder, thus xerces sax, this would prime
the same factory classes.  On closer review, the prime didn't even work
correctly, though it did load the classes.  At any rate, I was mostly
interested in why performance is the way it is rather than a head to head
comparison.  A difference of 40% doesn't match my own anecdotal evidence.


> - You GC before JDOM, but not DOM.

That was in the original and I missed that, but since I was not interested
in comparing memory and because the timer starts after gc, I don't know how
much it matters.  I updated the code and do see a difference, but still not
the kind of difference reported.  I am still comparing the deferred xerces
dom against the full JDOM model and now see a ~20% difference in favor of
Xerces FWIW.  This test is not nearly enough to go on but provided a simple
way to feed my trace files to see where the time is spent.

-------------- next part --------------
A non-text attachment was scrubbed...
Name: Compare.java
Type: application/octet-stream
Size: 3158 bytes
Desc: not available
Url : http://jdom.org/pipermail/jdom-interest/attachments/20010509/3851ed1b/Compare.obj


More information about the jdom-interest mailing list