[smokeping-users] The Smoke
Thomas Klein
smokeping at thomas-klein.de
Thu Feb 11 13:32:11 CET 2010
Thanks Chris and Tobias!
I have setup my table in a misleading way. Each Val column was meant as
one measurement of 10 values numbered from 1-10. This columns are
already sorted and the median should be 5.
Is there an easy way for me to create example graphs with such values to
be able to explain this to other people based on pictures?
Greetings,
Thomas
Tobias Oetiker wrote:
> Hi Thomas,
>
> Today Thomas Klein wrote:
>
>
>> Hello,
>>
>> is there a more detailed explanation available how SmokePing does draw the
>> smoke into the graphs?
>> I have looked into "Reading the Graphs" from the documentation, but miss some
>> details:
>>
>> For every round of measurement smokeping sends several packets. It then sorts
>> the different round trip times and selects the median, (ie. the middle one).
>> This means when there are 10 time values, value number 5 is selected and
>> drawn. The other values are drawn as successively lighter shades of gray in
>> the background (smoke).
>>
>> How can i find out what the smoke will look for such example values where the
>> median will be always 5?
>>
>
> the median is NOT always 5 ... the values get sorted and then the
> middle one is picked ... the medians would look like this:
>
>
>> Nr. Val1 Val2 Val3 Val4 Val5 median
>> 1 1 1 5 4 1 1
>> 2 2 2 5 5 1 2
>> 3 3 5 5 5 1 5
>> 4 4 5 5 5 1 5
>> 5 5 5 5 5 5 5
>> 6 6 5 5 5 5 5
>> 7 7 5 5 9 5 5
>> 8 8 5 5 9 5 5
>> 9 9 5 9 9 5 9
>> 10 10 5 10 9 5 9
>>
>
>
>
>> I have noted that my latency measurements tend to have more smoke above the
>> median measurement than below and wonder what this does mean and how to
>> interprete the graphs.
>>
>
> if more smoke is above than below (which is normally the case for
> ping latencies) that means that every now an then a packet is
> takes longer than most of the others ... but no packet is faster
> than any of the others.
>
> cheers
> tobi
>
>
>> Greetings,
>> Thomas
>>
>>
>
>
More information about the smokeping-users
mailing list