|
From: Rob H. <ro...@ca...> - 2004-11-29 17:09:15
|
All, I ran these tests on my machine and did a little tweaking with the parameters to see what performance improvements could be made - here is what I got. With the default configuration my figures were: 1008;945;968;929;;1109;1109;1039;1125;922;945;929;1179; With the proxyTargetClass flasg set to true I got: 617;570;593;578;;765;719;687;726;547;546;539;765; And with both frozen and proxyTargetClass set to true I got: 422;258;257;258;;437;406;343;414;234;234;234;453; This is the output given by the tool and each field is nanoseconds per invocation for a given test. As you can see it is more than possible to substantially improve the performance shown for Spring in the performance test suite. However, in the current version of Spring you cannot set the frozen when using the ProxyFactoryBean. I have a fix for this that I used for the test which involves overriding the setFrozen method on ProxyFactoryBean to set another flag instead of frozen and then using this flag to set the frozen flag just before proxy creation. If everyone is happy with this fix I will commit it to CVS. Rob Rob Harrop wrote: > All, > > An interesting article today about AOP performance and the new > performance suite created by the AspectWerkz team. As always I take > these with a pinch of salt but I am trying to get hold of the code so > that I can run the tests myself and see if the Spring figures can be > tweaked at all. > > http://www.theserverside.com/news/thread.tss?thread_id=30238 > > Rob > > > ------------------------------------------------------- > SF email is sponsored by - The IT Product Guide > Read honest & candid reviews on hundreds of IT Products from real users. > Discover which products truly live up to the hype. Start reading now. > http://productguide.itmanagersjournal.com/ > _______________________________________________ > Springframework-developer mailing list > Spr...@li... > https://lists.sourceforge.net/lists/listinfo/springframework-developer > > |