Hi all, I have a question. What method do you use to benchmark your webapps? I'm currently using "ab" (apache benchmark) but I'd like to be able to do comparative analysis.
Specifically, I'm testing POI::HSSF (big shock) Serializer in Cocoon. I'd like comparative numbers for the HSSF:Serializer against the HTML serializer. The numbers "ab" gives me are not shall we say believable. One moment it says the HSSF::Serializer (Excel file format) is FASTER then the HTML Serializer. Which I find very doubtful (no offense to our serializer)...sometimes it doesn't. The numbers vary greatly. Now I believe the thought that the numbers should very some->a lot but a rational test based on controlled parameters should give me some idea of a speed benchmark with some level of variance. The idea is to use this as a general point of reference for proving future builds/releases and concepts. While we do this easily with the APIs via junit tests and other "known" situations, this doesn't help us much with the serializer in a concurrent light/medium and heavy load situation. I'm not getting paid for this so neither should my benchmarking tool ;-). Thanks, Andy -- www.superlinksoftware.com www.sourceforge.net/projects/poi - port of Excel format to java http://developer.java.sun.com/developer/bugParade/bugs/4487555.html - fix java generics! The avalanche has already started. It is too late for the pebbles to vote. -Ambassador Kosh
