Hi! marc fleury wrote: > change the start() thingy to a real thread pool that takes the invocations > from teh clients (instead of starting a new server each time, should help > you ;-). As previous post, run() does not call start(). The setup is only done once. And besides, that's not what causing the performance problems. The client now caches the stub, so this is only done once. It's the calls to the stub that are slow. /Rickard -- Rickard �berg @home: +46 13 177937 Email: [EMAIL PROTECTED] http://www.telkel.com http://www.jboss.org http://www.dreambean.com
- [jBoss-Dev] RMI call latency to naming server=excessive Rickard �berg
- Re: [jBoss-Dev] RMI call latency to naming server=exc... Juha-P Lindfors
- Re: [jBoss-Dev] RMI call latency to naming server... Rickard �berg
- RE: [jBoss-Dev] RMI call latency to naming se... marc fleury
- RE: [jBoss-Dev] RMI call latency to namin... marc fleury
- RE: [jBoss-Dev] RMI call latency to ... Rickard �berg
- RE: [jBoss-Dev] RMI call latency... marc fleury
- Re: [jBoss-Dev] RMI call lat... Rickard �berg
- Re: [jBoss-Dev] RMI call latency to namin... Rickard �berg
- Re: [jBoss-Dev] RMI call latency to naming server=exc... Justin Forder
- Re: [jBoss-Dev] RMI call latency to naming server... Rickard �berg
- Re: [jBoss-Dev] RMI call latency to naming se... Justin Forder
