[xquery-talk] performance figures for xquery processors

[Makoto YUI]

Attention: 
	This is Java specific issue.

"Michael Kay" <mhk at mhk.me.uk> wrote:

> It's been suggested on this thread that the Saxon tree format takes more
> space than for example DTM. I would be surprised if there were a significant
> difference for most cases, so I'd be interested to see the evidence.

I should say the grounds..

While my test code for Saxon is missing by replacing previous 
with latest Saxon ;-), but test code for Qizx is still exist 
(attached).

Please examine the same test on Saxon and Qizx, if you mind.

The tiny (JUnit) test try to load XMark documents (scaling factor 
0.1/1, 10M/100M) and show memory usage and elapsed time.
# please replace path to files for your env.

While I can't say whether it is most cases,
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
the difference causes when treating large documents.

In my laptop, Saxon can't load 100M xml document.
it is because memory space (total 768M/free about 250M) is 
insufficient, but it'll scaling topic. That was why I put 
the issue.

Object creation is still heavy even in latest JVM 
and each Object has at least 8 bytes! memory footprint, 
this costs.

If you are using C++ for implimentation language, 
the difference would be so little.

I have to say additionally, I like Saxon comparatively.

Thanks,

Makoto

--
Nara Institute of Science and Technology (NAIST)
Makoto YUI <yuin405 AT gmail.com>
http://db-www.naist.jp/~makoto-y/proj/xbird/
-------------- next part --------------
A non-text attachment was scrubbed...
Name: DocumentParserTest.java
Type: application/octet-stream
Size: 1993 bytes
Desc: not available
Url : http://xquery.com/pipermail/talk/attachments/20060208/9dabc95a/DocumentParserTest.obj