G'day Java samplers ie custom built to the Java sampler specification. Basically the samplers send SOAP requests (but they do a lot more than that in preparing the SOAP statement and accounting for it afterwards).
I also was expecting (without having done any research) that JMeter would run at least as fast under the 1.5 JVM. Preserving performance of 1.4.2 (and ideally earlier) code would probably have been a design goal for 1.5. But I wouldn't have been surprised if I had to at least rebuild JMeter with Java 1.5 libraries to maintain performance. However I don't want to introduce rebuilding JMeter into the ongoing process here. So I'll keep my samplers at 1.4.2 especially now that I have redeveloped the necessary multithreading code to 1.4.2. Ian Blavins Contract Performance Engineer Temenos -----Original Message----- From: Peter Lin [mailto:[EMAIL PROTECTED] Sent: 25 April 2007 00:11 To: JMeter Users List Subject: Re: Performance of JMeter under Java 5 that's odd. Back in 2005, I did some tests and jmeter ran faster with jdk1.5.0 for http sampler for me. That was with a single jmeter instance, not distributed testing. what sampler are you using? peter On 4/24/07, sebb <[EMAIL PROTECTED]> wrote: > > I've no idea why that should be. > > There are no current plans to release a version of JMeter compiled for > Java 5. > > And I'd be surprised if merely recompiling fixed the performance hit. > Perhaps the JVM is less efficient? > > S > On 24/04/07, iblavins <[EMAIL PROTECTED]> wrote: > > G'day > > > > > > > > I changed the JVM under which I am running JMeter server from 1.4.2 to 5 > and > > found a reduction in throughput of about 25%. > > > > > > > > This was: > > > > - in distributed mode with the client on Windows XP and one > server > > on AIX > > > > - without rebuilding JMeter for Java 5 > > > > - with the client always at 1.5 > > > > - with the memory management options in the JMeter proc > commented > > out (for other reasons) but with plenty of memory allocated to the JVM > on > > the server > > > > - running JMeter hard but not excessively so (360 concurrent > users > > running flat chat but with 25% spare CPU on the server) > > > > - on a five minute run > > > > > > > > Is that consistent with expectations and if so is there a known way to > claw > > some of that performance back. > > > > > > > > (I'd prefer to run JMeter under 1.5 so my Java samplers can be at 1.5. > But > > for the moment at least I can't live with a 25% throughput loss.) > > > > > > > > > > > > > > > > > > > > Ian Blavins > > > > Contract Performance Engineer > > > > Temenos > > > > > > > > > > > > > > This email (and any attachments) contains confidential information, and > is intended > > only for the named recipient. Distribution or copying of this email by > anyone > > other than the named recipient is prohibited. If you are not the named > or > > intended recipient, please notify TEMENOS or the sender immediately and > > permanently destroy this email (and any attachments) and all copies of > it. No > > member of TEMENOS Group AG or any of its associated or affiliated > companies is > > liable for any errors or omissions in the content or transmission of > this > > email. Any opinions contained in this email are solely those of the > author and, > > unless clearly indicated otherwise in writing, are not endorsed by any > member > > of TEMENOS Group AG or any of its associated and affiliated companies. > > > > --------------------------------------------------------------------- > To unsubscribe, e-mail: [EMAIL PROTECTED] > For additional commands, e-mail: [EMAIL PROTECTED] > > This email (and any attachments) contains confidential information, and is intended only for the named recipient. Distribution or copying of this email by anyone other than the named recipient is prohibited. If you are not the named or intended recipient, please notify TEMENOS or the sender immediately and permanently destroy this email (and any attachments) and all copies of it. No member of TEMENOS Group AG or any of its associated or affiliated companies is liable for any errors or omissions in the content or transmission of this email. Any opinions contained in this email are solely those of the author and, unless clearly indicated otherwise in writing, are not endorsed by any member of TEMENOS Group AG or any of its associated and affiliated companies. --------------------------------------------------------------------- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
