You are viewing a plain text version of this content. The canonical link for it is here.
Posted to user@jmeter.apache.org by maalamaal <ve...@hotmail.com> on 2008/04/30 00:35:58 UTC
how do you analyse the results?
in my test case i am using no: of users as 60 and ramp up period as 60 and
loop count forever.
the response time i get how do i measure it and tell my client?
please help me as i am new to jmeter
--
View this message in context: http://www.nabble.com/how-do-you-analyse-the-results--tp16971969p16971969.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
Re: how do you analyse the results?
Posted by sebb <se...@gmail.com>.
2008/4/30 David Brown <da...@davidwbrown.name>:
> I have hit this limitation before brought to my attention by: Sebb. There is a concurrency issue (you can look it up in bugzilla) with multiple users per ThreadGroup. The work-a-round is to create several ThreadGroups with one user per ThreadGroup.
Huh?
I'm not aware ever recommending single-user threadgroups.
I think there must have been a misunderstanding.
> Ideally: you should use distributed master/slave testing as indicated in the link I sent you stated below. Ostensibly, the plan is if you can garner several machines (anything will do: laptops, old towers etc.) to be used as the slaves you are in business if you follow the doco link provided below. HTH.
>
Distributed testing is inherently more resource intensive for the
client system and the network.
> maalamaal wrote ..
>
>
> >
> > just one thread group
> > what happens if i have multiple thread groups?
> >
> >
> >
> > geezenslaw wrote:
> > >
> > > How many ThreadGroups are you using? This might help:
> > > http://jakarta.apache.org/jmeter/usermanual/jmeter_distributed_testing_step_by_step.pdf.
> > > HTH.
> > >
> > > maalamaal wrote ..
> > >>
> > >> in my test case i am using no: of users as 60 and ramp up period as 60
> > >> and
> > >> loop count forever.
> > >> the response time i get how do i measure it and tell my client?
> > >> please help me as i am new to jmeter
> > >> --
> > >> View this message in context:
> > >> http://www.nabble.com/how-do-you-analyse-the-results--tp16971969p16971969.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
> > > Yet some, not wise, go to the other side of the globe, to barbarous and
> > > unhealthy regions, and devote ten or twenty years, in that they may
> > > live,-that is, keep comfortably warm,- and die in New England at last.
> > >
> > > Henry David Thoreau - Walden - 1845
> > >
> > >
> > > ---------------------------------------------------------------------
> > > 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-do-you-analyse-the-results--tp16971969p16972476.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
> Yet some, not wise, go to the other side of the globe, to barbarous and unhealthy regions, and devote ten or twenty years, in that they may live,-that is, keep comfortably warm,- and die in New England at last.
>
> Henry David Thoreau - Walden - 1845
>
>
> ---------------------------------------------------------------------
> 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 do you analyse the results?
Posted by David Brown <da...@davidwbrown.name>.
Results analysis can be somewhat subjective based upon the metrics you deem important. Like all engineering Software engineering is not always Blank and White. From a software engineering perspective I tend to establish a baseline metric such as configuring a single ThreadGroup with a loop count of 1 with a Listener: View Results in Table with the following column label fields: Sample #, Start Time (in thousandths of a second), Label (application link url executed), Sample Time (ms), Status, # of Bytes. I then create another Test Plan with more than one ThreadGroup. Each ThreadGroup will have a loop count greater than 1, say, 3 for example. I then run the new Test Plan and then perform a so-called Gap Analysis which is a simple comparison between the the two different Test Plans used for the target report analysis. This means to compare the results of the baseline metric Test Plan to the second Test Plan that has a more aggressive load test built into the ThreadGroups. You co
uld also add another Listener such as a Graph Results Listener which will display graphically the results of your View in Table Results Listener. And, as stated below you could migrate this basic (baseline versus failure) testing model to a distributed Test Plan that can stress load test your system much more aggressively. The test model remains the same: 1. construct and execute a baseline Test Plan using arbitrarily low inputs (ThreadGroup users, loop count, etc.), 2. construct a failure Test Plan such that unit increments in ThreadGroup users and loop counts lead to application failure. Application failure means the first observable transaction loss. This means the target application failed to complete a transaction or dropped the intended transaction in total. Once both Test Plans have been completed through to gathering the tabulated and graphical results you can include screenshots of the baseline Test Plan compared to the results of the failure Test Plan in your targe
t report. HTH.
maalamaal wrote ..
>
> I wanted to know how do you document the result and explain it to the client
> based on response time.
>
>
>
> geezenslaw wrote:
> >
> > I have hit this limitation before brought to my attention by: Sebb. There
> > is a concurrency issue (you can look it up in bugzilla) with multiple
> > users per ThreadGroup. The work-a-round is to create several ThreadGroups
> > with one user per ThreadGroup. Ideally: you should use distributed
> > master/slave testing as indicated in the link I sent you stated below.
> > Ostensibly, the plan is if you can garner several machines (anything will
> > do: laptops, old towers etc.) to be used as the slaves you are in business
> > if you follow the doco link provided below. HTH.
> >
> > maalamaal wrote ..
> >>
> >> just one thread group
> >> what happens if i have multiple thread groups?
> >>
> >>
> >>
> >> geezenslaw wrote:
> >> >
> >> > How many ThreadGroups are you using? This might help:
> >> >
> >> http://jakarta.apache.org/jmeter/usermanual/jmeter_distributed_testing_step_by_step.pdf.
> >> > HTH.
> >> >
> >> > maalamaal wrote ..
> >> >>
> >> >> in my test case i am using no: of users as 60 and ramp up period as 60
> >> >> and
> >> >> loop count forever.
> >> >> the response time i get how do i measure it and tell my client?
> >> >> please help me as i am new to jmeter
> >> >> --
> >> >> View this message in context:
> >> >>
> >> http://www.nabble.com/how-do-you-analyse-the-results--tp16971969p16971969.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
> >> > Yet some, not wise, go to the other side of the globe, to barbarous and
> >> > unhealthy regions, and devote ten or twenty years, in that they may
> >> > live,-that is, keep comfortably warm,- and die in New England at last.
> >> >
> >> > Henry David Thoreau - Walden - 1845
> >> >
> >> >
> >> > ---------------------------------------------------------------------
> >> > 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-do-you-analyse-the-results--tp16971969p16972476.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
> > Yet some, not wise, go to the other side of the globe, to barbarous and
> > unhealthy regions, and devote ten or twenty years, in that they may
> > live,-that is, keep comfortably warm,- and die in New England at last.
> >
> > Henry David Thoreau - Walden - 1845
> >
> >
> > ---------------------------------------------------------------------
> > 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-do-you-analyse-the-results--tp16971969p16972763.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
Yet some, not wise, go to the other side of the globe, to barbarous and unhealthy regions, and devote ten or twenty years, in that they may live,-that is, keep comfortably warm,- and die in New England at last.
Henry David Thoreau - Walden - 1845
---------------------------------------------------------------------
To unsubscribe, e-mail: jmeter-user-unsubscribe@jakarta.apache.org
For additional commands, e-mail: jmeter-user-help@jakarta.apache.org
Re: how do you analyse the results?
Posted by maalamaal <ve...@hotmail.com>.
I wanted to know how do you document the result and explain it to the client
based on response time.
geezenslaw wrote:
>
> I have hit this limitation before brought to my attention by: Sebb. There
> is a concurrency issue (you can look it up in bugzilla) with multiple
> users per ThreadGroup. The work-a-round is to create several ThreadGroups
> with one user per ThreadGroup. Ideally: you should use distributed
> master/slave testing as indicated in the link I sent you stated below.
> Ostensibly, the plan is if you can garner several machines (anything will
> do: laptops, old towers etc.) to be used as the slaves you are in business
> if you follow the doco link provided below. HTH.
>
> maalamaal wrote ..
>>
>> just one thread group
>> what happens if i have multiple thread groups?
>>
>>
>>
>> geezenslaw wrote:
>> >
>> > How many ThreadGroups are you using? This might help:
>> >
>> http://jakarta.apache.org/jmeter/usermanual/jmeter_distributed_testing_step_by_step.pdf.
>> > HTH.
>> >
>> > maalamaal wrote ..
>> >>
>> >> in my test case i am using no: of users as 60 and ramp up period as 60
>> >> and
>> >> loop count forever.
>> >> the response time i get how do i measure it and tell my client?
>> >> please help me as i am new to jmeter
>> >> --
>> >> View this message in context:
>> >>
>> http://www.nabble.com/how-do-you-analyse-the-results--tp16971969p16971969.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
>> > Yet some, not wise, go to the other side of the globe, to barbarous and
>> > unhealthy regions, and devote ten or twenty years, in that they may
>> > live,-that is, keep comfortably warm,- and die in New England at last.
>> >
>> > Henry David Thoreau - Walden - 1845
>> >
>> >
>> > ---------------------------------------------------------------------
>> > 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-do-you-analyse-the-results--tp16971969p16972476.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
> Yet some, not wise, go to the other side of the globe, to barbarous and
> unhealthy regions, and devote ten or twenty years, in that they may
> live,-that is, keep comfortably warm,- and die in New England at last.
>
> Henry David Thoreau - Walden - 1845
>
>
> ---------------------------------------------------------------------
> 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-do-you-analyse-the-results--tp16971969p16972763.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
Re: how do you analyse the results?
Posted by David Brown <da...@davidwbrown.name>.
I have hit this limitation before brought to my attention by: Sebb. There is a concurrency issue (you can look it up in bugzilla) with multiple users per ThreadGroup. The work-a-round is to create several ThreadGroups with one user per ThreadGroup. Ideally: you should use distributed master/slave testing as indicated in the link I sent you stated below. Ostensibly, the plan is if you can garner several machines (anything will do: laptops, old towers etc.) to be used as the slaves you are in business if you follow the doco link provided below. HTH.
maalamaal wrote ..
>
> just one thread group
> what happens if i have multiple thread groups?
>
>
>
> geezenslaw wrote:
> >
> > How many ThreadGroups are you using? This might help:
> > http://jakarta.apache.org/jmeter/usermanual/jmeter_distributed_testing_step_by_step.pdf.
> > HTH.
> >
> > maalamaal wrote ..
> >>
> >> in my test case i am using no: of users as 60 and ramp up period as 60
> >> and
> >> loop count forever.
> >> the response time i get how do i measure it and tell my client?
> >> please help me as i am new to jmeter
> >> --
> >> View this message in context:
> >> http://www.nabble.com/how-do-you-analyse-the-results--tp16971969p16971969.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
> > Yet some, not wise, go to the other side of the globe, to barbarous and
> > unhealthy regions, and devote ten or twenty years, in that they may
> > live,-that is, keep comfortably warm,- and die in New England at last.
> >
> > Henry David Thoreau - Walden - 1845
> >
> >
> > ---------------------------------------------------------------------
> > 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-do-you-analyse-the-results--tp16971969p16972476.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
Yet some, not wise, go to the other side of the globe, to barbarous and unhealthy regions, and devote ten or twenty years, in that they may live,-that is, keep comfortably warm,- and die in New England at last.
Henry David Thoreau - Walden - 1845
---------------------------------------------------------------------
To unsubscribe, e-mail: jmeter-user-unsubscribe@jakarta.apache.org
For additional commands, e-mail: jmeter-user-help@jakarta.apache.org
Re: how do you analyse the results?
Posted by maalamaal <ve...@hotmail.com>.
just one thread group
what happens if i have multiple thread groups?
geezenslaw wrote:
>
> How many ThreadGroups are you using? This might help:
> http://jakarta.apache.org/jmeter/usermanual/jmeter_distributed_testing_step_by_step.pdf.
> HTH.
>
> maalamaal wrote ..
>>
>> in my test case i am using no: of users as 60 and ramp up period as 60
>> and
>> loop count forever.
>> the response time i get how do i measure it and tell my client?
>> please help me as i am new to jmeter
>> --
>> View this message in context:
>> http://www.nabble.com/how-do-you-analyse-the-results--tp16971969p16971969.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
> Yet some, not wise, go to the other side of the globe, to barbarous and
> unhealthy regions, and devote ten or twenty years, in that they may
> live,-that is, keep comfortably warm,- and die in New England at last.
>
> Henry David Thoreau - Walden - 1845
>
>
> ---------------------------------------------------------------------
> 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-do-you-analyse-the-results--tp16971969p16972476.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
Re: how do you analyse the results?
Posted by David Brown <da...@davidwbrown.name>.
How many ThreadGroups are you using? This might help: http://jakarta.apache.org/jmeter/usermanual/jmeter_distributed_testing_step_by_step.pdf. HTH.
maalamaal wrote ..
>
> in my test case i am using no: of users as 60 and ramp up period as 60 and
> loop count forever.
> the response time i get how do i measure it and tell my client?
> please help me as i am new to jmeter
> --
> View this message in context: http://www.nabble.com/how-do-you-analyse-the-results--tp16971969p16971969.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
Yet some, not wise, go to the other side of the globe, to barbarous and unhealthy regions, and devote ten or twenty years, in that they may live,-that is, keep comfortably warm,- and die in New England at last.
Henry David Thoreau - Walden - 1845
---------------------------------------------------------------------
To unsubscribe, e-mail: jmeter-user-unsubscribe@jakarta.apache.org
For additional commands, e-mail: jmeter-user-help@jakarta.apache.org