With 2.0.3 and earlier, the rate is calculated *per thread*, so you need to allow for this. The documentation should say this, but doesn't.
[The forthcoming 2.1 has options to do the calculation for all threads etc.] S. On 7/5/05, Martin McAtamney <[EMAIL PROTECTED]> wrote: > Hi, > > > > New to JMeter, so be gentle. > > > > I have a very simple test. I want to hit the home page of my application > with 40 requests per second. I am using remote testing, with jmeter-server > on three UNIX environments, and just running the test from my windows box. > There doesn't appear to be an issue with generating the load, but I am > having problems in controlling the load, i.e. getting it steady at 40 > requests per second. > > > > The approach I took was a Thread Group with 200 threads, a http request just > to hit the home page, and a Constant Throughput timer with a target through > put of 12 samples a minute. > > > > My theory was 200 threads, (users) * 12 iterations per min = 2400 request > per min, i.e. 40 request per sec. However the test appears to be generating > a lot more that 40 request per second. > > > > Any ideas on what might be wrong, or have I missed something. > > > > > > Any help/advice appreciated > > > > M. > > > > > > > --------------------------------------------------------------------- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
