[XML-SIG] re : XSL Performance / Processors

[Daniel Veillard]

On Tue, Apr 01, 2003 at 05:06:36PM +0200, Philippe Dubreuil wrote:
> But nobody can explain my results :
> "Performance can depend on many apparently unrelated factors (including
> the fluttering of the wings of a butterfly...)" M.Kay 
[...]
> Some of my results (execution Time)  (RedHat 6.2, i686, 256Mo RAM, PIII 
> 550Mhz)
> I'm using the same structure (index or not) three times in this example.
> If I use 10 times, i get the same results.
> 
>  
>                   withIndex               noIndex
> Xalan2.5             8780ms                 9108ms    
> Saxon6.5.2  tree     2800ms                 2900ms
> "          tinytree 3021ms                 1757ms   !!!!
> 

  If you want to do performance timing, using a Java based implementation
is the best way to introduce irrationality in your results. The effects
of the HotSpot algorithm detection and of the asynchronous garbage collection
simply makes any test run less than 50 times a joke, because they don't
represent the output of the real working set nor show GC effects.
  In general the more you pile up complex layers, the hardest it is
to get predicatble results for the whole stack,

   have fun !

Daniel

-- 
Daniel Veillard      | Red Hat Network https://rhn.redhat.com/
veillard@redhat.com  | libxml GNOME XML XSLT toolkit  http://xmlsoft.org/
http://veillard.com/ | Rpmfind RPM search engine http://rpmfind.net/