cocoon-users mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Berin Loritsch <blorit...@apache.org>
Subject Re: Relative Speeds of XSL Engines
Date Thu, 01 Feb 2001 14:44:21 GMT
Arnaud Vandyck wrote:
> 
> ----- Original Message -----
> From: "Timm, Sean" <STimm@mailgo.com>
> To: <cocoon-users@xml.apache.org>
> Sent: Wednesday, January 31, 2001 8:37 PM
> Subject: RE: Relative Speeds of XSL Engines
> 
> > I'm not sure that these benchmarks can be completely trusted with
> > such a small sampling of test files (three, to be exact).  I'd
> > be interested in seeing a comparison that utilized a wide variety
> > of XSL tests and multiple JVMs.  It looks like
> > http://www.datapower.com/XSLTMark/ has a more thorough test suite,
> > but they don't have benchmarks posted for Xalan2.
> 
> I'm not sure I really understand these tests! I always hear "Java is slow"
> and in these tests, Java XSLT Processor are sometimes faster than C++
> parsers! Am I right? Did I miss something?.. or maybe missunderstand?
>

You didn't miss anything.  Most of the issue is algorithm.
A slow algorithm is slow no matter what language you throw
at it.  For the same exact algorithm, java will run slower
than C++.  But you can't do the exact same algorithm most of
the time which leads to optimizations depending on your
implementation language.

Mime
View raw message