Subject: RE: [xsl] measuring bulk performance & turn around times of XSL t ransformations? ideas for: XML to XML, XML to HTML, XML to FO (then to P D F) From: Pieter Reint Siegers Kort <pieter.siegers@xxxxxxxxxxx> Date: Wed, 28 Apr 2004 10:16:43 -0500 |
Of course Kevin, but once it is loaded and given large input, the transformation times were quite useful in telling me what coding technique was the fastest and which had the smallest use of memory. I obtained the memory-in-use information from the available performance counters. Basically, all the (simple) testing told me that just by increasing machine power (CPU, RAM) I solved most problems, as the coding techniques were very close to each other. This doesn't mean that I wouldn't be interested in setting up a much more reliable performance testing application or tool - like you're talking about, but most of the time, a developer's time is very limited :-) <prs/> -----Original Message----- From: Kevin Jones [mailto:kjones@xxxxxxxxxxx] Sent: Tuesday, April 27, 2004 4:01 PM To: xsl-list@xxxxxxxxxxxxxxxxxxxxxx Subject: Re: [xsl] measuring bulk performance & turn around times of XSL t ransformations? ideas for: XML to XML, XML to HTML, XML to FO (then to PD F) On Tuesday 27 April 2004 6:50 pm, you wrote: > > Utilities like XSelerator show you the time needed to do when doing > XML parse, XSL parse, and XSL Transformation time. Very useful. I used > various techniques to process XML and compared those times using MSXML > 4.0 (you can use other processors by simply adding them to the Environmental options). I would be very careful about only doing that. There are a lot of startup costs in most XSLT processors that cause problems with one shots tests. You almost certainly need to use a custom driver that matches how you would like to use the processors to get accurate figures. If its WIN32 only, the Microsoft recommendation used to be carry on using MSXML (via COM) as the .NET XSLT was still under development. Of the publicly available processors MSXML always comes very high up the benchmarks. If you want more performance you need to look at the companies that specialize in XML appliances and/or do some code tuning. I have spent a fair amount of time comparing XSLT processor performance so feel free to ask some detailed questions. I also have access to pretty much the full range of processors if there is some standalone bit of XSLT you are interested in getting figures for. Kev Sarvega Inc.
Current Thread |
---|
|
<- Previous | Index | Next -> |
---|---|---|
RE: [xsl] XSL Problem, Kenny Akridge | Thread | Re: [xsl] measuring bulk performanc, Kevin Jones |
[xsl] XSL Problem, Andrea Racca | Date | RE: [xsl] measuring bulk performanc, Pieter Reint Siegers |
Month |