[152647] in North American Network Operators' Group
Re: Network diagram app that shows realtime link utilizatin
daemon@ATHENA.MIT.EDU (Livio Zanol Puppim)
Fri May 4 22:46:35 2012
In-Reply-To: <4FA49052.2000104@tiggee.com>
Date: Fri, 4 May 2012 23:45:57 -0300
From: Livio Zanol Puppim <livio.zanol.puppim@gmail.com>
Cc: nanog@nanog.org
Errors-To: nanog-bounces+nanog.discuss=bloom-picayune.mit.edu@nanog.org
In a troubleshooting situation I think real time is valid. Off course, is
non-sense doing this for "normal" monitoring...
But firstly define: What's real-time? every second? miliseconds? 10
seconds? 30 seconds? It's subjective nowadays
2012/5/4 David Miller <dmiller@tiggee.com>
> On 5/4/2012 6:53 PM, Jeroen van Aart wrote:
> > Anurag Bhatia wrote:
> >> I have been using Zenoss quite a bit. It does not shows exact real tim=
e
> >> stat of interface but close to real time + it has ton more options for
> >
> > I remember someone here saying that real time monitoring gives you
> > useless results, because if you make the time of measurement small
> > enough the utilisation becomes 100%. Measurement of throughput is how
> > many times this happens during a particular time interval. I hope I
> > remembered right.
> >
>
> I think you are referring to this thread -
> http://www.gossamer-threads.com/lists/nanog/users/149903
>
> and in particular this quote:
> cjp at 0x1 wrote on Feb 16, 2012, 11:25am
> As sampling rate approaches zero, so will the "spikyness" of the
> graph--ultimately an interface is either sending a frame (*100*%) or
> it's
> not (0%).
>
> -cjp
>
>
> Utilisation doesn't "become 100%", instead measured utilisation will
> either be 100% or 0% at each interval.
>
> -DMM
>
>
>
--=20
[]'s
L=EDvio Zanol Puppim
