You are viewing a plain text version of this content. The canonical link for it is here.
Posted to user@jmeter.apache.org by Thomas Johnson <nt...@gmail.com> on 2009/07/01 01:54:53 UTC
Re: how to setup this scenario?
I can't offer much by way of configuration suggestions, but just some
napkin math to determine what the simulated user behavior should be
like:
30,000 iterations per hour / 10 concurrent users = 3,000 iterations of
10 threads
3000 iterations per hour / 60 mins per hour = 50 iterations per minute
50 iterations per minute / 60 seconds per minute = 1.2 seconds per iteration
1.2 seconds per iteration / 7 pages per iteration = ~0.2 seconds per page
Assuming that you're testing the user-facing website part of things,
rather than a web service, so you are effectively asking users to take
0.2 seconds for each page view, including images and loading time.
You'll likely want to slow that down somewhat, by increasing the
number of concurrent users (say 1000 instead of 10) to give closer to
20 seconds per page.
30000/1000 = 30 iterations per thread
1/(30 / 60 / 60) = 120 seconds per iteration
120/7 = 17 seconds per page
If I may ask, would it be possible to get further details on what your
testing goals are? It's not quite clear what scenario you are trying
to test for. Namely, is the goal to test:
- 30,000 unique sessions per hour
- 10 users for an hour (casual use)
- 10 users for an hour (clicking madly every 0.2 seconds)
- 30,000 users for an hour
Each of these tests represents a very different scenario, so it would
help to get a better idea of exactly what you are trying to model,
then to configure JMeter based upon that.
On Tue, Jun 30, 2009 at 11:01 PM, olafmolilink<ol...@ilink.nl> wrote:
>
> Hello,
>
> i have to test a website for a peakload of 30.000 users during a period of 1
> hour, with 10 concurrent users that request 7 webpages each.
> How can i best setup this scenario in Jmeter? Creating 30.000 threads
> doesn't seem like the way to go, but i am stuck thinking of another way..
>
> Olaf
> --
> View this message in context: http://www.nabble.com/how-to-setup-this-scenario--tp24279552p24279552.html
> Sent from the JMeter - User mailing list archive at Nabble.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: how to setup this scenario?
Posted by olafmolilink <ol...@ilink.nl>.
Hello Thomas,
thanks for your reply. The scenario would be " - 30,000 unique sessions per
hour".
All these unique sessions (visitors) would request about 7 webpages, and we
want to see how the server responds, and how long each webpage take to
return to the client. We are aiming at a max. of 2seconds response time for
each page.
Olaf
Thomas Johnson-8 wrote:
>
> I can't offer much by way of configuration suggestions, but just some
> napkin math to determine what the simulated user behavior should be
> like:
>
> 30,000 iterations per hour / 10 concurrent users = 3,000 iterations of
> 10 threads
> 3000 iterations per hour / 60 mins per hour = 50 iterations per minute
> 50 iterations per minute / 60 seconds per minute = 1.2 seconds per
> iteration
> 1.2 seconds per iteration / 7 pages per iteration = ~0.2 seconds per page
>
> Assuming that you're testing the user-facing website part of things,
> rather than a web service, so you are effectively asking users to take
> 0.2 seconds for each page view, including images and loading time.
> You'll likely want to slow that down somewhat, by increasing the
> number of concurrent users (say 1000 instead of 10) to give closer to
> 20 seconds per page.
>
> 30000/1000 = 30 iterations per thread
> 1/(30 / 60 / 60) = 120 seconds per iteration
> 120/7 = 17 seconds per page
>
> If I may ask, would it be possible to get further details on what your
> testing goals are? It's not quite clear what scenario you are trying
> to test for. Namely, is the goal to test:
> - 30,000 unique sessions per hour
> - 10 users for an hour (casual use)
> - 10 users for an hour (clicking madly every 0.2 seconds)
> - 30,000 users for an hour
>
> Each of these tests represents a very different scenario, so it would
> help to get a better idea of exactly what you are trying to model,
> then to configure JMeter based upon that.
>
> On Tue, Jun 30, 2009 at 11:01 PM, olafmolilink<ol...@ilink.nl> wrote:
>>
>> Hello,
>>
>> i have to test a website for a peakload of 30.000 users during a period
>> of 1
>> hour, with 10 concurrent users that request 7 webpages each.
>> How can i best setup this scenario in Jmeter? Creating 30.000 threads
>> doesn't seem like the way to go, but i am stuck thinking of another way..
>>
>> Olaf
>> --
>> View this message in context:
>> http://www.nabble.com/how-to-setup-this-scenario--tp24279552p24279552.html
>> Sent from the JMeter - User mailing list archive at Nabble.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
>
>
>
--
View this message in context: http://www.nabble.com/how-to-setup-this-scenario--tp24279552p24284268.html
Sent from the JMeter - User mailing list archive at Nabble.com.
---------------------------------------------------------------------
To unsubscribe, e-mail: jmeter-user-unsubscribe@jakarta.apache.org
For additional commands, e-mail: jmeter-user-help@jakarta.apache.org