jmeter-user mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Deepak Shetty <shet...@gmail.com>
Subject Re: Test plan for (970 page requests every 5 min)
Date Mon, 20 Sep 2010 20:09:52 GMT
>So the bottom line is that in order to calculate a page loading time using
>JMeter, a tester needs to find out the proportion using different set ups
>and then use that proportion to interpret the test data. Correct?
Yes with some caveats as usual.
a. this is my opinion, use it if it works for you or makes sense to you.
Felix for e.g. has a different perspective and you might find something
different that makes life easier for you. i can afford to do what I do
because I have a psychical architecture that supports it.
b. very accurate page load times are rarely needed . you have limited amount
of time and resources so run the type of tests that give you more value for
your time/money.
c. Different browsers (and the network) behave differently , so even this is
an estimate. For e.g. before we tuned our last website users in CA would get
a response time of 4 seconds v/s users in NY who got 7-11 seconds. But youll
really wont have the time to test all possible combinations and variations.
Getting a sense of your times and using some sort of safety factor is
usually enough(but my managers are technical so they understand the
problems).

>What else would you suggest to test on a static pages (no dynamically
generated stuff)?
I would spend more time on the dynamically generated stuff since that's
where problems crop up.

>If response time for all resources increase with the same proportion, it is
uniformly.
If memory serves me correctly you had a load , but the times for the same
resource varied wildly. if this is not the case ignore what I said. If the
problem is server CPU then this should be easy to verify, run perfmon(or
equivalent) while your test is running and check if the CPU is pegged at
100% on the server.

>What is a JMeter source? Is it another log file?
I meant the source code.


On Mon, Sep 20, 2010 at 12:19 PM, Prostak <strodion@gmail.com> wrote:

>
> >No Im saying in the specific case that you really need accurate response
> >times and the architecture is like yours , you might need to do this.
> >Again there are so many factors that influence a page load time as a
> browser
> >sees that it is not usually worth doing any  exhaustive testing for this.
>
> So the bottom line is that in order to calculate a page loading time using
> JMeter, a tester needs to find out the proportion using different set ups
> and then use that proportion to interpret the test data. Correct?
>
> >(And page load time is just one part of a family of load testing)
>
> What else would you suggest to test on a static pages (no dynamically
> generated stuff)?
>
> >In some of your original screens the times for the same static file varied
> >from a few ms to a few seconds. Thats not uniformly slow. If your Server
> was
> >loaded , then it should be simple to find this out using perfmon
>
> If response time for all resources increase with the same proportion, it is
> uniformly.
>
> >Yes but you get a java stack trace that may have some more information or
> >lets you look into JMeter source and see what it was trying to do.
>
> What is a JMeter source? Is it another log file?
> --
> View this message in context:
> http://jmeter.512774.n5.nabble.com/Test-plan-for-970-page-requests-every-5-min-tp2826174p2847080.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
>
>

Mime
  • Unnamed multipart/alternative (inline, None, 0 bytes)
View raw message