Matt, Currently we just do prototyping with the stuff, no profiling at all. I´m away three weeks for holiday. Let us Discuss this when I come back, ok? It´s crucial and your measures should help a lot in that respect.
CGJ -----Ursprüngliche Nachricht----- Von: Matt Munz [mailto:mmunz@;apelon.com] Gesendet: Freitag, 1. November 2002 18:14 An: [EMAIL PROTECTED] Betreff: RE: [JBoss-dev] JBoss.net and performance tuning hi again, Anybody out there working on or using JBoss.net? Attached are some files I'm using to test, in a very simplistic way, the performance of jmx.net. My current timing so far is on average .3 seconds to complete the simplest possible transaction (as far as I can tell) -- returning a very short string object. Does this number sound reasonable? If so, what are you using JBoss.Net for? Perhaps I have the wrong idea... Not included in the archive is the class JmxNetProxy, which is a delegate for the mechanism used in the JBoss.Net test cases. Here's the relevant method from that class. public Object invoke(String webServiceName, String mbeanServiceName, String methodName, Object[] arguments, Class[] classes) throws Exception { ... MBeanInvocationHandler handler; handler = createMBeanInvocationHandler(target); return handler.invoke(mbeanServiceName, methodName, arguments, classes); // classes } Any comment would be greatly appreciated. - Matt ([EMAIL PROTECTED]) -----Original Message----- From: [EMAIL PROTECTED] [mailto:jboss-development-admin@;lists.sourceforge.net]On Behalf Of Matt Munz Sent: Thursday, October 31, 2002 11:17 AM To: JBoss Developers Group Subject: [JBoss-dev] JBoss.net and performance tuning CGJ and JBoss.net guys, My fledgeling JBoss.net enabled application is growing up and it needs some performance enhancements. Particularly, Marshalling/Unmarshalling appears to take significantly longer than expected. Here's my setup. 1) I'm using the JMX.net proxy classes used in the test cases. 2) I am working with short, frequent transactions. 3) I am using the bean serializer to send simple custom objects across the wire. 4) On the server side, an MBean is the recipient of the request. RE: #1 -- Would WSDL2Java be any faster? Is MBean to wsdl generation working yet? RE: #2 -- I know, course-grained transactions are preferred, but are they required? My objects are small, almost tiny. How fast can a transaction be? If I can get it under .05 seconds, this should suffice for the moment. RE: #3 -- Any tips / tricks here? Would switiching to primitives and Strings be markedly faster? Any help would be greatly appreciated. Links to benchmarks / performance tests would help too. - Matt ------------------------------------------------------- This sf.net email is sponsored by: Influence the future of Java(TM) technology. Join the Java Community Process(SM) (JCP(SM)) program now. http://ads.sourceforge.net/cgi-bin/redirect.pl?sunm0004en _______________________________________________ Jboss-development mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/jboss-development ########################################### This message has been scanned by F-Secure Anti-Virus for Microsoft Exchange. For more information, connect to http://www.F-Secure.com/ ------------------------------------------------------- This sf.net email is sponsored by: See the NEW Palm Tungsten T handheld. Power & Color in a compact size! http://ads.sourceforge.net/cgi-bin/redirect.pl?palm0001en _______________________________________________ Jboss-development mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/jboss-development
