You are viewing a plain text version of this content. The canonical link for it is here.
Posted to user@jmeter.apache.org by indireddysankar <in...@gmail.com> on 2009/05/15 11:07:59 UTC

How to understand My graph results

Hi,
        I have ceated my test plan as shown below

T(50)
R(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 ,
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

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...

   Thanks in  advance...
-Siva

Re: How to understand My graph results

Posted by sebb <se...@gmail.com>.
On 15/05/2009, indireddysankar <in...@gmail.com> 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 <se...@gmail.com> wrote:
>
>  > On 15/05/2009, indireddysankar <in...@gmail.com> 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: jmeter-user-unsubscribe@jakarta.apache.org
>  > For additional commands, e-mail: jmeter-user-help@jakarta.apache.org
>  >
>  >
>

---------------------------------------------------------------------
To unsubscribe, e-mail: jmeter-user-unsubscribe@jakarta.apache.org
For additional commands, e-mail: jmeter-user-help@jakarta.apache.org


Re: How to understand My graph results

Posted by indireddysankar <in...@gmail.com>.
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

On Fri, May 15, 2009 at 3:16 PM, sebb <se...@gmail.com> wrote:

> On 15/05/2009, indireddysankar <in...@gmail.com> 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: jmeter-user-unsubscribe@jakarta.apache.org
> For additional commands, e-mail: jmeter-user-help@jakarta.apache.org
>
>

Re: How to understand My graph results

Posted by sebb <se...@gmail.com>.
On 15/05/2009, indireddysankar <in...@gmail.com> 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: jmeter-user-unsubscribe@jakarta.apache.org
For additional commands, e-mail: jmeter-user-help@jakarta.apache.org