You are viewing a plain text version of this content. The canonical link for it is here.
Posted to user@jmeter.apache.org by Sebastiano Spicuglia <sp...@gmail.com> on 2013/09/26 15:24:31 UTC
Summariser throughput calculation
Hello all,
I have a question about the throughput calculation in Jmeter.
I set up a simple test plan and I configured jmeter to print information
about the throughput using the information found in:
http://jmeter.apache.org/usermanual/component_reference.html#Generate_Summary_Results
The results I got are the following:
summary + 1 in 0.2s = 4.6/s Avg: 218 Min: 218 Max: 218
Err: 0 (0.00%) Active: 3 Started: 3 Finished: 0
summary + 92 in 5.3s = 17.2/s Avg: 784 Min: 199 Max: 1409
Err: 0 (0.00%) Active: 34 Started: 34 Finished: 0
summary = 93 in 5.3s = 17.4/s Avg: 778 Min: 199 Max: 1409
Err: 0 (0.00%)
summary + 113 in 7.3s = 15.4/s Avg: 2068 Min: 1413 Max: 2638
Err: 0 (0.00%) Active: 60 Started: 60 Finished: 0
summary = 206 in 11.3s = 18.2/s Avg: 1485 Min: 199 Max: 2638
Err: 0 (0.00%)
summary + 130 in 9s = 14.7/s Avg: 2829 Min: 2461 Max: 3128
Err: 0 (0.00%) Active: 60 Started: 60 Finished: 0
summary = 336 in 17.4s = 19.3/s Avg: 2005 Min: 199 Max: 3128
Err: 0 (0.00%)
summary + 123 in 9s = 13.8/s Avg: 2878 Min: 2574 Max: 3082
Err: 0 (0.00%) Active: 60 Started: 60 Finished: 0
summary = 459 in 23.3s = 19.7/s Avg: 2239 Min: 199 Max: 3128
Err: 0 (0.00%)
summary + 127 in 9s = 14.3/s Avg: 2873 Min: 2554 Max: 3200
Err: 0 (0.00%) Active: 60 Started: 60 Finished: 0
summary = 586 in 29.3s = 20.0/s Avg: 2376 Min: 199 Max: 3200
Err: 0 (0.00%)
summary + 125 in 9s = 14.0/s Avg: 2876 Min: 2614 Max: 3090
Err: 0 (0.00%) Active: 60 Started: 60 Finished: 0
summary = 711 in 35.3s = 20.1/s Avg: 2464 Min: 199 Max: 3200
Err: 0 (0.00%)
I am a little bit lost here.
Why is the sum of all the intermediate intervals higher than final time?
If I did the math right the sum should be 48.8 while the final interval is 35.3.
Why is the final throughput (that should be the cumulative) higher
than all the intermediate?
Could you please explain me how to read those results?
Thank you very much for you support and congratulation for your job
Best regards
Sebastiano
---------------------------------------------------------------------
To unsubscribe, e-mail: user-unsubscribe@jmeter.apache.org
For additional commands, e-mail: user-help@jmeter.apache.org
Re: Summariser throughput calculation
Posted by Thatos Chronos <th...@gmail.com>.
I'm curious, what can you use for formal reporting of JMeter run outputs?
Many thanks,
Jon.
On 27 September 2013 12:52, sebb <se...@gmail.com> wrote:
> On 27 September 2013 07:27, Sebastiano Spicuglia
> <sp...@gmail.com> wrote:
> > Hi,
> >
> > thank you for the reply.
>
> What is the JMeter version?
>
> Are you running the test in stand-alone or client-server mode?
>
> > Sorry I'm quite new with JMeter, where could I find the
> > timestamp of the readings?
>
> In the jmeter.log file
>
> > Btw: I checked what is the maximum throughput of that server
> > (the same info I was trying to collect with JMeter) with another load
> > generator (httperf). It is around 20req/s.
> > Therefore the last line of Jmeter's log seems to report a reasonable
> > number.
> >
> > Is it possible that the cumulative and the delta are computed
> > totally independently?
>
> No, the same data is used.
>
> >
> > - Sebastiano
> >
> > On Thu, Sep 26, 2013 at 5:52 PM, sebb <se...@gmail.com> wrote:
> >> On 26 September 2013 14:24, Sebastiano Spicuglia
> >> <sp...@gmail.com> wrote:
> >>> Hello all,
> >>>
> >>> I have a question about the throughput calculation in Jmeter.
> >>
> >> Which version of JMeter?
> >>
> >>> I set up a simple test plan and I configured jmeter to print
> information
> >>> about the throughput using the information found in:
> >>>
> >>>
> http://jmeter.apache.org/usermanual/component_reference.html#Generate_Summary_Results
> >>>
> >>> The results I got are the following:
> >>>
> >>> summary + 1 in 0.2s = 4.6/s Avg: 218 Min: 218 Max: 218
> >>> Err: 0 (0.00%) Active: 3 Started: 3 Finished: 0
> >>> summary + 92 in 5.3s = 17.2/s Avg: 784 Min: 199 Max: 1409
> >>> Err: 0 (0.00%) Active: 34 Started: 34 Finished: 0
> >>> summary = 93 in 5.3s = 17.4/s Avg: 778 Min: 199 Max: 1409
> >>> Err: 0 (0.00%)
> >>> summary + 113 in 7.3s = 15.4/s Avg: 2068 Min: 1413 Max: 2638
> >>> Err: 0 (0.00%) Active: 60 Started: 60 Finished: 0
> >>> summary = 206 in 11.3s = 18.2/s Avg: 1485 Min: 199 Max: 2638
> >>> Err: 0 (0.00%)
> >>> summary + 130 in 9s = 14.7/s Avg: 2829 Min: 2461 Max: 3128
> >>> Err: 0 (0.00%) Active: 60 Started: 60 Finished: 0
> >>> summary = 336 in 17.4s = 19.3/s Avg: 2005 Min: 199 Max: 3128
> >>> Err: 0 (0.00%)
> >>> summary + 123 in 9s = 13.8/s Avg: 2878 Min: 2574 Max: 3082
> >>> Err: 0 (0.00%) Active: 60 Started: 60 Finished: 0
> >>> summary = 459 in 23.3s = 19.7/s Avg: 2239 Min: 199 Max: 3128
> >>> Err: 0 (0.00%)
> >>> summary + 127 in 9s = 14.3/s Avg: 2873 Min: 2554 Max: 3200
> >>> Err: 0 (0.00%) Active: 60 Started: 60 Finished: 0
> >>> summary = 586 in 29.3s = 20.0/s Avg: 2376 Min: 199 Max: 3200
> >>> Err: 0 (0.00%)
> >>> summary + 125 in 9s = 14.0/s Avg: 2876 Min: 2614 Max: 3090
> >>> Err: 0 (0.00%) Active: 60 Started: 60 Finished: 0
> >>> summary = 711 in 35.3s = 20.1/s Avg: 2464 Min: 199 Max: 3200
> >>> Err: 0 (0.00%)
> >>>
> >>> I am a little bit lost here.
> >>>
> >>> Why is the sum of all the intermediate intervals higher than final
> time?
> >>
> >> Rounding may account for some of the discrepancy, but it does seem odd
> >> that the final few intermediates are 9s apart whereas the totals are
> >> about 6s apart.
> >>
> >> Can you provide the timestamps for the readings?
> >>
> >>> If I did the math right the sum should be 48.8 while the final
> interval is 35.3.
> >>>
> >>> Why is the final throughput (that should be the cumulative) higher
> >>> than all the intermediate?
> >>
> >> The throughput is samples/time so if the times are different the
> >> throughput will also be different.
> >>
> >>> Could you please explain me how to read those results?
> >>
> >> The '+' lines are current interval only and the '=' are cumulative.
> >>
> >> The output is intended as a guide to how the test is progressing, not
> >> for formal reporting.
> >>
> >>> Thank you very much for you support and congratulation for your job
> >>> Best regards
> >>> Sebastiano
> >>>
> >>> ---------------------------------------------------------------------
> >>> To unsubscribe, e-mail: user-unsubscribe@jmeter.apache.org
> >>> For additional commands, e-mail: user-help@jmeter.apache.org
> >>>
> >>
> >> ---------------------------------------------------------------------
> >> To unsubscribe, e-mail: user-unsubscribe@jmeter.apache.org
> >> For additional commands, e-mail: user-help@jmeter.apache.org
> >>
> >
> > ---------------------------------------------------------------------
> > To unsubscribe, e-mail: user-unsubscribe@jmeter.apache.org
> > For additional commands, e-mail: user-help@jmeter.apache.org
> >
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: user-unsubscribe@jmeter.apache.org
> For additional commands, e-mail: user-help@jmeter.apache.org
>
>
Re: Summariser throughput calculation
Posted by sebb <se...@gmail.com>.
On 27 September 2013 07:27, Sebastiano Spicuglia
<sp...@gmail.com> wrote:
> Hi,
>
> thank you for the reply.
What is the JMeter version?
Are you running the test in stand-alone or client-server mode?
> Sorry I'm quite new with JMeter, where could I find the
> timestamp of the readings?
In the jmeter.log file
> Btw: I checked what is the maximum throughput of that server
> (the same info I was trying to collect with JMeter) with another load
> generator (httperf). It is around 20req/s.
> Therefore the last line of Jmeter's log seems to report a reasonable
> number.
>
> Is it possible that the cumulative and the delta are computed
> totally independently?
No, the same data is used.
>
> - Sebastiano
>
> On Thu, Sep 26, 2013 at 5:52 PM, sebb <se...@gmail.com> wrote:
>> On 26 September 2013 14:24, Sebastiano Spicuglia
>> <sp...@gmail.com> wrote:
>>> Hello all,
>>>
>>> I have a question about the throughput calculation in Jmeter.
>>
>> Which version of JMeter?
>>
>>> I set up a simple test plan and I configured jmeter to print information
>>> about the throughput using the information found in:
>>>
>>> http://jmeter.apache.org/usermanual/component_reference.html#Generate_Summary_Results
>>>
>>> The results I got are the following:
>>>
>>> summary + 1 in 0.2s = 4.6/s Avg: 218 Min: 218 Max: 218
>>> Err: 0 (0.00%) Active: 3 Started: 3 Finished: 0
>>> summary + 92 in 5.3s = 17.2/s Avg: 784 Min: 199 Max: 1409
>>> Err: 0 (0.00%) Active: 34 Started: 34 Finished: 0
>>> summary = 93 in 5.3s = 17.4/s Avg: 778 Min: 199 Max: 1409
>>> Err: 0 (0.00%)
>>> summary + 113 in 7.3s = 15.4/s Avg: 2068 Min: 1413 Max: 2638
>>> Err: 0 (0.00%) Active: 60 Started: 60 Finished: 0
>>> summary = 206 in 11.3s = 18.2/s Avg: 1485 Min: 199 Max: 2638
>>> Err: 0 (0.00%)
>>> summary + 130 in 9s = 14.7/s Avg: 2829 Min: 2461 Max: 3128
>>> Err: 0 (0.00%) Active: 60 Started: 60 Finished: 0
>>> summary = 336 in 17.4s = 19.3/s Avg: 2005 Min: 199 Max: 3128
>>> Err: 0 (0.00%)
>>> summary + 123 in 9s = 13.8/s Avg: 2878 Min: 2574 Max: 3082
>>> Err: 0 (0.00%) Active: 60 Started: 60 Finished: 0
>>> summary = 459 in 23.3s = 19.7/s Avg: 2239 Min: 199 Max: 3128
>>> Err: 0 (0.00%)
>>> summary + 127 in 9s = 14.3/s Avg: 2873 Min: 2554 Max: 3200
>>> Err: 0 (0.00%) Active: 60 Started: 60 Finished: 0
>>> summary = 586 in 29.3s = 20.0/s Avg: 2376 Min: 199 Max: 3200
>>> Err: 0 (0.00%)
>>> summary + 125 in 9s = 14.0/s Avg: 2876 Min: 2614 Max: 3090
>>> Err: 0 (0.00%) Active: 60 Started: 60 Finished: 0
>>> summary = 711 in 35.3s = 20.1/s Avg: 2464 Min: 199 Max: 3200
>>> Err: 0 (0.00%)
>>>
>>> I am a little bit lost here.
>>>
>>> Why is the sum of all the intermediate intervals higher than final time?
>>
>> Rounding may account for some of the discrepancy, but it does seem odd
>> that the final few intermediates are 9s apart whereas the totals are
>> about 6s apart.
>>
>> Can you provide the timestamps for the readings?
>>
>>> If I did the math right the sum should be 48.8 while the final interval is 35.3.
>>>
>>> Why is the final throughput (that should be the cumulative) higher
>>> than all the intermediate?
>>
>> The throughput is samples/time so if the times are different the
>> throughput will also be different.
>>
>>> Could you please explain me how to read those results?
>>
>> The '+' lines are current interval only and the '=' are cumulative.
>>
>> The output is intended as a guide to how the test is progressing, not
>> for formal reporting.
>>
>>> Thank you very much for you support and congratulation for your job
>>> Best regards
>>> Sebastiano
>>>
>>> ---------------------------------------------------------------------
>>> To unsubscribe, e-mail: user-unsubscribe@jmeter.apache.org
>>> For additional commands, e-mail: user-help@jmeter.apache.org
>>>
>>
>> ---------------------------------------------------------------------
>> To unsubscribe, e-mail: user-unsubscribe@jmeter.apache.org
>> For additional commands, e-mail: user-help@jmeter.apache.org
>>
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: user-unsubscribe@jmeter.apache.org
> For additional commands, e-mail: user-help@jmeter.apache.org
>
---------------------------------------------------------------------
To unsubscribe, e-mail: user-unsubscribe@jmeter.apache.org
For additional commands, e-mail: user-help@jmeter.apache.org
Re: Summariser throughput calculation
Posted by Sebastiano Spicuglia <sp...@gmail.com>.
Hi,
thank you for the reply.
Sorry I'm quite new with JMeter, where could I find the
timestamp of the readings?
Btw: I checked what is the maximum throughput of that server
(the same info I was trying to collect with JMeter) with another load
generator (httperf). It is around 20req/s.
Therefore the last line of Jmeter's log seems to report a reasonable
number.
Is it possible that the cumulative and the delta are computed
totally independently?
- Sebastiano
On Thu, Sep 26, 2013 at 5:52 PM, sebb <se...@gmail.com> wrote:
> On 26 September 2013 14:24, Sebastiano Spicuglia
> <sp...@gmail.com> wrote:
>> Hello all,
>>
>> I have a question about the throughput calculation in Jmeter.
>
> Which version of JMeter?
>
>> I set up a simple test plan and I configured jmeter to print information
>> about the throughput using the information found in:
>>
>> http://jmeter.apache.org/usermanual/component_reference.html#Generate_Summary_Results
>>
>> The results I got are the following:
>>
>> summary + 1 in 0.2s = 4.6/s Avg: 218 Min: 218 Max: 218
>> Err: 0 (0.00%) Active: 3 Started: 3 Finished: 0
>> summary + 92 in 5.3s = 17.2/s Avg: 784 Min: 199 Max: 1409
>> Err: 0 (0.00%) Active: 34 Started: 34 Finished: 0
>> summary = 93 in 5.3s = 17.4/s Avg: 778 Min: 199 Max: 1409
>> Err: 0 (0.00%)
>> summary + 113 in 7.3s = 15.4/s Avg: 2068 Min: 1413 Max: 2638
>> Err: 0 (0.00%) Active: 60 Started: 60 Finished: 0
>> summary = 206 in 11.3s = 18.2/s Avg: 1485 Min: 199 Max: 2638
>> Err: 0 (0.00%)
>> summary + 130 in 9s = 14.7/s Avg: 2829 Min: 2461 Max: 3128
>> Err: 0 (0.00%) Active: 60 Started: 60 Finished: 0
>> summary = 336 in 17.4s = 19.3/s Avg: 2005 Min: 199 Max: 3128
>> Err: 0 (0.00%)
>> summary + 123 in 9s = 13.8/s Avg: 2878 Min: 2574 Max: 3082
>> Err: 0 (0.00%) Active: 60 Started: 60 Finished: 0
>> summary = 459 in 23.3s = 19.7/s Avg: 2239 Min: 199 Max: 3128
>> Err: 0 (0.00%)
>> summary + 127 in 9s = 14.3/s Avg: 2873 Min: 2554 Max: 3200
>> Err: 0 (0.00%) Active: 60 Started: 60 Finished: 0
>> summary = 586 in 29.3s = 20.0/s Avg: 2376 Min: 199 Max: 3200
>> Err: 0 (0.00%)
>> summary + 125 in 9s = 14.0/s Avg: 2876 Min: 2614 Max: 3090
>> Err: 0 (0.00%) Active: 60 Started: 60 Finished: 0
>> summary = 711 in 35.3s = 20.1/s Avg: 2464 Min: 199 Max: 3200
>> Err: 0 (0.00%)
>>
>> I am a little bit lost here.
>>
>> Why is the sum of all the intermediate intervals higher than final time?
>
> Rounding may account for some of the discrepancy, but it does seem odd
> that the final few intermediates are 9s apart whereas the totals are
> about 6s apart.
>
> Can you provide the timestamps for the readings?
>
>> If I did the math right the sum should be 48.8 while the final interval is 35.3.
>>
>> Why is the final throughput (that should be the cumulative) higher
>> than all the intermediate?
>
> The throughput is samples/time so if the times are different the
> throughput will also be different.
>
>> Could you please explain me how to read those results?
>
> The '+' lines are current interval only and the '=' are cumulative.
>
> The output is intended as a guide to how the test is progressing, not
> for formal reporting.
>
>> Thank you very much for you support and congratulation for your job
>> Best regards
>> Sebastiano
>>
>> ---------------------------------------------------------------------
>> To unsubscribe, e-mail: user-unsubscribe@jmeter.apache.org
>> For additional commands, e-mail: user-help@jmeter.apache.org
>>
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: user-unsubscribe@jmeter.apache.org
> For additional commands, e-mail: user-help@jmeter.apache.org
>
---------------------------------------------------------------------
To unsubscribe, e-mail: user-unsubscribe@jmeter.apache.org
For additional commands, e-mail: user-help@jmeter.apache.org
Re: Summariser throughput calculation
Posted by sebb <se...@gmail.com>.
On 26 September 2013 14:24, Sebastiano Spicuglia
<sp...@gmail.com> wrote:
> Hello all,
>
> I have a question about the throughput calculation in Jmeter.
Which version of JMeter?
> I set up a simple test plan and I configured jmeter to print information
> about the throughput using the information found in:
>
> http://jmeter.apache.org/usermanual/component_reference.html#Generate_Summary_Results
>
> The results I got are the following:
>
> summary + 1 in 0.2s = 4.6/s Avg: 218 Min: 218 Max: 218
> Err: 0 (0.00%) Active: 3 Started: 3 Finished: 0
> summary + 92 in 5.3s = 17.2/s Avg: 784 Min: 199 Max: 1409
> Err: 0 (0.00%) Active: 34 Started: 34 Finished: 0
> summary = 93 in 5.3s = 17.4/s Avg: 778 Min: 199 Max: 1409
> Err: 0 (0.00%)
> summary + 113 in 7.3s = 15.4/s Avg: 2068 Min: 1413 Max: 2638
> Err: 0 (0.00%) Active: 60 Started: 60 Finished: 0
> summary = 206 in 11.3s = 18.2/s Avg: 1485 Min: 199 Max: 2638
> Err: 0 (0.00%)
> summary + 130 in 9s = 14.7/s Avg: 2829 Min: 2461 Max: 3128
> Err: 0 (0.00%) Active: 60 Started: 60 Finished: 0
> summary = 336 in 17.4s = 19.3/s Avg: 2005 Min: 199 Max: 3128
> Err: 0 (0.00%)
> summary + 123 in 9s = 13.8/s Avg: 2878 Min: 2574 Max: 3082
> Err: 0 (0.00%) Active: 60 Started: 60 Finished: 0
> summary = 459 in 23.3s = 19.7/s Avg: 2239 Min: 199 Max: 3128
> Err: 0 (0.00%)
> summary + 127 in 9s = 14.3/s Avg: 2873 Min: 2554 Max: 3200
> Err: 0 (0.00%) Active: 60 Started: 60 Finished: 0
> summary = 586 in 29.3s = 20.0/s Avg: 2376 Min: 199 Max: 3200
> Err: 0 (0.00%)
> summary + 125 in 9s = 14.0/s Avg: 2876 Min: 2614 Max: 3090
> Err: 0 (0.00%) Active: 60 Started: 60 Finished: 0
> summary = 711 in 35.3s = 20.1/s Avg: 2464 Min: 199 Max: 3200
> Err: 0 (0.00%)
>
> I am a little bit lost here.
>
> Why is the sum of all the intermediate intervals higher than final time?
Rounding may account for some of the discrepancy, but it does seem odd
that the final few intermediates are 9s apart whereas the totals are
about 6s apart.
Can you provide the timestamps for the readings?
> If I did the math right the sum should be 48.8 while the final interval is 35.3.
>
> Why is the final throughput (that should be the cumulative) higher
> than all the intermediate?
The throughput is samples/time so if the times are different the
throughput will also be different.
> Could you please explain me how to read those results?
The '+' lines are current interval only and the '=' are cumulative.
The output is intended as a guide to how the test is progressing, not
for formal reporting.
> Thank you very much for you support and congratulation for your job
> Best regards
> Sebastiano
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: user-unsubscribe@jmeter.apache.org
> For additional commands, e-mail: user-help@jmeter.apache.org
>
---------------------------------------------------------------------
To unsubscribe, e-mail: user-unsubscribe@jmeter.apache.org
For additional commands, e-mail: user-help@jmeter.apache.org