You are viewing a plain text version of this content. The canonical link for it is here.
Posted to user@jmeter.apache.org by Joshua Auerbach <jo...@dealer.com> on 2007/12/06 20:22:42 UTC
Summary Report - Throughput Question
I am encountering some strange behavior when looking at the throughput
column of the summary report with multiple remote machines. Hoping
someone could clarify this for me
For example when I have one remote machine and run my test plan with two
threads (each request having a random delay of 1-2s and the test plan
set to run forever) I am seeing my total throughput stabilize around
1.2/sec which makes sense.
When I take the same test plan and bump it up to 4 threads still from
one remote machine I am seeing a throughput stabilize of about 2.2/sec
which also makes sense (being almost twice what I was seeing with 2 threads)
When I run this test plan on 2 remote machines each with 2 threads I
would expect to see similar results as running 4 threads on one machine
however what I am seeing is the throughput takes a much longer time to
stabilize and slowly works its way up from a very low number (showing up
as /hour) and never see it come close to the 2.2/sec I was getting from
one machine.
Any help would be appreciated.
Cheers,
Josh
--
______________________________
* Joshua E. Auerbach | Software Developer
V 802.658.0965
josh@dealer.com | www.dealer.com
*
--
______________________________
* Joshua E. Auerbach | Software Developer
V 802.658.0965
josh@dealer.com | www.dealer.com
*
---------------------------------------------------------------------
To unsubscribe, e-mail: jmeter-user-unsubscribe@jakarta.apache.org
For additional commands, e-mail: jmeter-user-help@jakarta.apache.org
Re: Summary Report - Throughput Question
Posted by sebb <se...@gmail.com>.
On 06/12/2007, Joshua Auerbach <jo...@dealer.com> wrote:
> All I see related to this on that page is "*Throughput * is calculated
> as requests/unit of time. The time is calculated from the start of the
> first sample to the end of the last sample. This includes any intervals
> between samples."
>
> What I don't understand is why the throughput would be drastically
> different between 4 threads spread over 2 machines vs 4 threads on a
> single machine (and moreover data that doesn't make sense whenever I
> have > 1 machine)
Are you sure that both server machines can generate the same load on their own?
It would be worth checking.
Also, try running the tests using non-GUI mode and then combining the
throughput calculations.
I think it's unlikely that using 2 remote servers instead of 1, but
with the same load would cause problems for client-server mode,
however it's worth checking.
> -Josh
>
> sebb wrote:
> > http://jakarta.apache.org/jmeter/usermanual/glossary.html may help
> >
> > On 06/12/2007, Joshua Auerbach <jo...@dealer.com> wrote:
> >
> >> I am encountering some strange behavior when looking at the throughput
> >> column of the summary report with multiple remote machines. Hoping
> >> someone could clarify this for me
> >>
> >> For example when I have one remote machine and run my test plan with two
> >> threads (each request having a random delay of 1-2s and the test plan
> >> set to run forever) I am seeing my total throughput stabilize around
> >> 1.2/sec which makes sense.
> >>
> >> When I take the same test plan and bump it up to 4 threads still from
> >> one remote machine I am seeing a throughput stabilize of about 2.2/sec
> >> which also makes sense (being almost twice what I was seeing with 2 threads)
> >>
> >> When I run this test plan on 2 remote machines each with 2 threads I
> >> would expect to see similar results as running 4 threads on one machine
> >> however what I am seeing is the throughput takes a much longer time to
> >> stabilize and slowly works its way up from a very low number (showing up
> >> as /hour) and never see it come close to the 2.2/sec I was getting from
> >> one machine.
> >>
> >> Any help would be appreciated.
> >>
> >> Cheers,
> >> Josh
> >> --
> >> ______________________________
> >> * Joshua E. Auerbach | Software Developer
> >>
> >> V 802.658.0965
> >> josh@dealer.com | www.dealer.com
> >> *
> >> --
> >> ______________________________
> >> * Joshua E. Auerbach | Software Developer
> >>
> >> V 802.658.0965
> >> josh@dealer.com | www.dealer.com
> >> *
> >>
> >> ---------------------------------------------------------------------
> >> 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
> >
> >
>
> --
> ______________________________
> * Joshua E. Auerbach | Software Developer
>
> V 802.658.0965
> josh@dealer.com | www.dealer.com
> *
>
> ---------------------------------------------------------------------
> 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: Summary Report - Throughput Question
Posted by Joshua Auerbach <jo...@dealer.com>.
All I see related to this on that page is "*Throughput * is calculated
as requests/unit of time. The time is calculated from the start of the
first sample to the end of the last sample. This includes any intervals
between samples."
What I don't understand is why the throughput would be drastically
different between 4 threads spread over 2 machines vs 4 threads on a
single machine (and moreover data that doesn't make sense whenever I
have > 1 machine)
-Josh
sebb wrote:
> http://jakarta.apache.org/jmeter/usermanual/glossary.html may help
>
> On 06/12/2007, Joshua Auerbach <jo...@dealer.com> wrote:
>
>> I am encountering some strange behavior when looking at the throughput
>> column of the summary report with multiple remote machines. Hoping
>> someone could clarify this for me
>>
>> For example when I have one remote machine and run my test plan with two
>> threads (each request having a random delay of 1-2s and the test plan
>> set to run forever) I am seeing my total throughput stabilize around
>> 1.2/sec which makes sense.
>>
>> When I take the same test plan and bump it up to 4 threads still from
>> one remote machine I am seeing a throughput stabilize of about 2.2/sec
>> which also makes sense (being almost twice what I was seeing with 2 threads)
>>
>> When I run this test plan on 2 remote machines each with 2 threads I
>> would expect to see similar results as running 4 threads on one machine
>> however what I am seeing is the throughput takes a much longer time to
>> stabilize and slowly works its way up from a very low number (showing up
>> as /hour) and never see it come close to the 2.2/sec I was getting from
>> one machine.
>>
>> Any help would be appreciated.
>>
>> Cheers,
>> Josh
>> --
>> ______________________________
>> * Joshua E. Auerbach | Software Developer
>>
>> V 802.658.0965
>> josh@dealer.com | www.dealer.com
>> *
>> --
>> ______________________________
>> * Joshua E. Auerbach | Software Developer
>>
>> V 802.658.0965
>> josh@dealer.com | www.dealer.com
>> *
>>
>> ---------------------------------------------------------------------
>> 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
>
>
--
______________________________
* Joshua E. Auerbach | Software Developer
V 802.658.0965
josh@dealer.com | www.dealer.com
*
---------------------------------------------------------------------
To unsubscribe, e-mail: jmeter-user-unsubscribe@jakarta.apache.org
For additional commands, e-mail: jmeter-user-help@jakarta.apache.org
Re: Summary Report - Throughput Question
Posted by sebb <se...@gmail.com>.
http://jakarta.apache.org/jmeter/usermanual/glossary.html may help
On 06/12/2007, Joshua Auerbach <jo...@dealer.com> wrote:
> I am encountering some strange behavior when looking at the throughput
> column of the summary report with multiple remote machines. Hoping
> someone could clarify this for me
>
> For example when I have one remote machine and run my test plan with two
> threads (each request having a random delay of 1-2s and the test plan
> set to run forever) I am seeing my total throughput stabilize around
> 1.2/sec which makes sense.
>
> When I take the same test plan and bump it up to 4 threads still from
> one remote machine I am seeing a throughput stabilize of about 2.2/sec
> which also makes sense (being almost twice what I was seeing with 2 threads)
>
> When I run this test plan on 2 remote machines each with 2 threads I
> would expect to see similar results as running 4 threads on one machine
> however what I am seeing is the throughput takes a much longer time to
> stabilize and slowly works its way up from a very low number (showing up
> as /hour) and never see it come close to the 2.2/sec I was getting from
> one machine.
>
> Any help would be appreciated.
>
> Cheers,
> Josh
> --
> ______________________________
> * Joshua E. Auerbach | Software Developer
>
> V 802.658.0965
> josh@dealer.com | www.dealer.com
> *
> --
> ______________________________
> * Joshua E. Auerbach | Software Developer
>
> V 802.658.0965
> josh@dealer.com | www.dealer.com
> *
>
> ---------------------------------------------------------------------
> 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