On 15/05/2009, indireddysankar <[email protected]> wrote: > Thanks sebbaz, > > One more doubt ,If rampup perriod is nothing to do with TPS then how > do i calculate my Load > i.e. How much is the Load(threads per sec) at which i am hitting my > server...with the following plan... > > My understanding is 50 users(threads) accessing server > simultaneosly...meaning 50TPS...am i correct.. > > Please clarify
Think about how JMeter runs each thread. JMeter runs the sampler. The sampler makes the request, and gets the response. JMeter processes the next element in the test plan, which in this case is the sampler again (as there are no timers). So what determines when the server sees the next sample? > > On Fri, May 15, 2009 at 3:16 PM, sebb <[email protected]> wrote: > > > On 15/05/2009, indireddysankar <[email protected]> wrote: > > > Hi, > > > I have ceated my test plan as shown below > > > > > > T(50) > > > R(10) > > > > I assume you mean Threads=50 and Ramp-up=10? > > > > > Loops(3600) > > > HTTP Sampler > > > > > > I am getting graph results as > > > Average latency is 1985ms , Throughput as 22.9/sec .. > > > > > > My understanding is > > > we loading the server with Threads per second(users)(50/10=5) i.e. TPS > > =5 , > > > > No, the ramp-up has little or no bearing on the TPS, except possibly > > for the first iteration. > > It only controls how long JMeter takes to start all the threads, after > > that they are on their own. > > > > > so at max the server has to show me max throughput it is handling as > > 5tps, > > > since we are htting 5tps and if server is handling all requests > > successfully > > > then throughput has to be 5/sec. How come throughput is 22.9 > > > > Throughput is calculated as requests/elapsed time. > > > > > i have gone through > > > http://jakarta.apache.org/jmeter/usermanual/test_plan.html ... > > > can anyone please interpret these results for me, where i am missing... > > > > See above. > > > > > Thanks in advance... > > > -Siva > > > > > > > > --------------------------------------------------------------------- > > 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]
