Hi >so that i get quite the processing time when running the services plus jmeter on localhost. you'd almost never do this. Your CPU and memory is limited so running a load test from the same machine on which the application is hosted will usually give you extremely poor results. You can't eliminate the network if your responses are reasonably sized unless you measure your times on the server itself . Usually if you test on a high speed LAN and your total traffic is significantly less than the bandwidth then you can guesstimate the network time to subtract though this is rarely needed.
> I want to get the PURE PROCESSING TIME of the services itself Why?. Measure it at the server if this is so important (technology dependent measurement ) regards deepak On Tue, Nov 23, 2010 at 4:24 AM, Albrecht Weiser <[email protected]> wrote: > I see. Then i was nearly right. If i measure latency on localhost, there's > nearly no time for delivering the request and response, so that i get quite > the processing time when running the services plus jmeter on localhost. > Thanks a lot > Albrecht > > > At 23.11.2010 11:52, sebb wrote: > >> On 23 November 2010 10:33, Albrecht Weiser<[email protected]> wrote: >> >>> Hi Jmeter users, >>> i have created a Jmeter test scenario. I'm testing several Webservices >>> there. I want to get the PURE PROCESSING TIME of the services itself >>> (without network losses, etc.). Jmeter offers two times to get with the >>> listeners: >>> 1. Time needed (I think it's overall time) - t >>> 2. Latency time (?) - lt >>> If googleing, the statements to latency time are very general. From my >>> point >>> of view, i need the latency time, if the processing time of a service is >>> ment. >>> Can anyone verify my assumption? >>> >> http://jakarta.apache.org/jmeter/usermanual/glossary.html#Latency >> >> Bye >>> Albrecht >>> >>> -- >>> Albrecht Weiser >>> mailto:aweiser AT gmx.de >>> >>> >>> --------------------------------------------------------------------- >>> To unsubscribe, e-mail: [email protected] >>> For additional commands, e-mail: [email protected] >>> >>> >>> --------------------------------------------------------------------- >> To unsubscribe, e-mail: [email protected] >> For additional commands, e-mail: [email protected] >> >> >> > > --------------------------------------------------------------------- > To unsubscribe, e-mail: [email protected] > For additional commands, e-mail: [email protected] > >
