[43252] in North American Network Operators' Group
Gordius has left the building. Was: RE: The Gorgon's Knot.
daemon@ATHENA.MIT.EDU (Bender, Andrew)
Tue Oct 2 23:10:31 2001
MIME-Version: 1.0
Content-Type: text/plain;
charset="Windows-1252"
Content-Transfer-Encoding: quoted-printable
Date: Tue, 2 Oct 2001 23:09:50 -0400
content-class: urn:content-classes:message
Message-ID: <D730399C6D1CD411B6DC00508BAC05362A316B@apollo.taqua.com>
From: "Bender, Andrew" <abender@taqua.com>
To: <nanog@merit.edu>
Errors-To: owner-nanog-outgoing@merit.edu
Deepak Jain wrote:=20
> The problem is where the RIB, or forwarding table is exported to=20
> the line cards which frequently have less than 128MB of usable=20
> RAM/SRAM/memory storage/etc. This essentially means that the line=20
> cards can only directly talk to other line cards for a specified,=20
> limited number of routing prefixes. I do not know the algorithms=20
> used when the line card is out of memory, but in many cases this=20
> memory is not field upgradeable beyond a certain point.
Srinivasan, Vargese, and others have demonstrated / observed that space
is not a material problem for the architecture you describe. Rather, it
is the time to complete a longest-match (or equivalent) operation for
each forwarding event, especially for high(er) interface speeds.
Regardless, 128MB is a LOT of table memory, especially when it is
virtually unknown to see anything larger than 10-20Mbit CAMs; prior to
reduction / expansion / etc, there's scarcely 11B of unique information
in an IPv4 route. Adding in some bytes for mysterious 'other stuff',
this is easily 4-8MM routes. We'll have to see another knee in route
proliferation before this becomes a problem, and space exhaust would
most certainly prevent this from happening.=20
> Many smaller networks can and do use PCs for BGP and for=20
> forwarding because their total forwarding needs at their core are=20
> say sufficiently less than 800mb/s upto which PCs seem to handle.=20
> However, the desires and models of these smaller networks don't=20
> scale much beyond this level with currently available PC=20
> technology.
Agreed... as regards PCs lets remember that history has offered certain
commentary on the non-role of PCs (indeed general purpose computers) in
network infrastructure... viz. there aren't many GRFs / DDP516Rs / etc.
around anymore. I doubt this is an 'interesting problem' for the op or
planning community.=20
Regards,
Andrew Bender
taqua.com
