Re: XSL performance problem

[Scott_Boag/CAM/Lotus <CN=Scott_Boag/OU=CAM/O=Lotus@xxxxxxxxx>]

I think it's a bit unfair to be judging XSL performance at this time.  Both
XT and LotusXSL are documented to be not optimized yet.  There are a LOT of
things to be done for optimization, and I know both James and myself (and
other processors like XSL:P) have these optimizations in our plans.  I
think you'll be pleased at how fast XSL can become.  But right now
stability and draft supporting features come first.

Remember how early in the game it is -- XSL is not yet a recommended
standard, and the last draft brought radical change to the expression
syntax.

-scott





Sebastien Sahuc <ssahuc@xxxxxxxxxxxxxx>@mulberrytech.com on 05/18/99
06:19:08 AM

Please respond to xsl-list@xxxxxxxxxxxxxxxx

Sent by:  owner-xsl-list@xxxxxxxxxxxxxxxx


To:   xsl-list@xxxxxxxxxxxxxxxx
cc:
Subject:  XSL performance problem


Hi there,

To be direct, I'll say that i'm quite disapointed about XML + XSL
processing as they seem to work really slowly, at least too slow to be
integrated on a back end architecture in java ( just like Cocoon
project). Indeed, after evaluating several parser and XSL processors,
I found out that the couple XP + XT is the fastest, but still don't
give good performance :
In a Pentium III, 450Mhz with Apache + Jserv + Sun JVM 1.2, a xml
document of 100Kb size processed with a small xsl document (that
simply format data into list of items) take a little less than 1000ms.
I know there is nothing comparable to what I'm arguing, structure of
xml and xsl documents can sensibly affect the result, but I'm really
surprised not to get better results.

Does it surprise somebody else ?

Any repy will be greatly appreciate.

Sebastien Sahuc
ssahuc@xxxxxxxxxxxxxx


 XSL-List info and archive:  http://www.mulberrytech.com/xsl/xsl-list





 XSL-List info and archive:  http://www.mulberrytech.com/xsl/xsl-list