[149855] in North American Network Operators' Group
Re: Hi speed trading - hi speed monitoring
daemon@ATHENA.MIT.EDU (Christopher J. Pilkington)
Thu Feb 16 14:26:41 2012
Date: Thu, 16 Feb 2012 14:25:32 -0500
From: "Christopher J. Pilkington" <cjp@0x1.net>
To: Hank Nussbacher <hank@efes.iucc.ac.il>
In-Reply-To: <5.1.0.14.2.20120216150028.00c2df78@efes.iucc.ac.il>
Cc: nanog@nanog.org
Errors-To: nanog-bounces+nanog.discuss=bloom-picayune.mit.edu@nanog.org
On Thu, Feb 16, 2012 at 03:03:55PM +0200, Hank Nussbacher wrote:
> Anyone who has managed a network knows that when you look at your
> MRTG/Cacti graphs at 5min, 10min ,15min intervals - all looks well.
> Start looking at 1sec intervals and you will see spikes that hit
> 100% of capacity - even on networks running at 25% average
> utilization.
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
