[3411] in North American Network Operators' Group
Re: Ping flooding (fwd)
daemon@ATHENA.MIT.EDU (Dorian Kim)
Wed Jul 10 04:40:14 1996
Date: Wed, 10 Jul 1996 04:34:57 -0400 (EDT)
From: Dorian Kim <dorian@cic.net>
To: Sean Doran <smd@icp.net>
cc: curtis@ans.net, nanog@merit.edu
In-Reply-To: <96Jul10.004335+0100_edt.20636+4@chops.icp.net>
On Wed, 10 Jul 1996, Sean Doran wrote:
> The current path appears -- to me, qualitatively -- to consume
> less CPU per packet per second in 7500s than the other released
> switching methods. Yes, I was seriously surprised too, however
> People That Know with three letter logins weren't.
With about 32K active flows, I've seen the same Without hard numbers, I'd put
the performance gain at 5-10% CPU over other fast switching method, but then
again, that's a total eyeball estimate. Pretty amazing stuff.
> conf t
> flow-export <ip-address>
> ^Z
Minor nit.
ip flow-export <ip-address>
^^
:)
It's always nice to see new technology being introduced. Flow switching, along
with Nevin Brownlee's real time flow visualisation tools should allow us much
better view into what's going on in our networks, an always useful thing IMO.
-dorian
