[6592] in North American Network Operators' Group
Re: NAP/ISP Saturation WAS: Re: Exchanges that matter...
daemon@ATHENA.MIT.EDU (Curtis Villamizar)
Fri Dec 20 23:52:30 1996
To: "Alex.Bligh" <amb@xara.net>
cc: Tony Li <tli@jnx.com>, david@sparks.net, nanog@merit.edu
Reply-To: curtis@ans.net
In-reply-to: Your message of "Fri, 20 Dec 1996 10:28:32 GMT."
<199612201028.KAA21015@diamond.xara.net>
Date: Fri, 20 Dec 1996 23:42:29 -0500
From: Curtis Villamizar <curtis@ans.net>
In message <199612201028.KAA21015@diamond.xara.net>, "Alex.Bligh" writes:
>
> Well this is the real point isn't it. Though Cisco has obviously paid
> an awful lot of attention to switching packets fast its host IP stack
> seems to be particularly bad compared to something built with a decent
> TCP/IP stack. If Cisco don't implement ICMP response in some sort
> of kernel layer below "application" layers handling other such functions
> the obvious question is "why not?". It must be possible to do this - just
> take your Netstar and downgrade the processor to a 386SX and see if it
> still works (fast external switching engine, slow processor).
Worked fine with an RS/6000 instead of the 586. Just generate the
ICMP on the forwarding card. It has a CPU on it, so use it. The
overhead of sending to the main processor, then back, then usually out
the same interface is probably much higher than just building the
packet and sticking it on the output ring buffer.
Curtis
