[88188] in North American Network Operators' Group
Re: Split flows across Domains
daemon@ATHENA.MIT.EDU (Robert E.Seastrom)
Tue Jan 24 13:14:27 2006
To: Joe Abley <jabley@isc.org>
Cc: "Christopher L. Morrow" <christopher.morrow@verizonbusiness.com>,
Glen Kent <glen.kent@gmail.com>, NANOG list <nanog@merit.edu>
From: Robert E.Seastrom <rs@seastrom.com>
Date: Tue, 24 Jan 2006 13:09:08 -0500
In-Reply-To: <3DFF80EF-2BDE-4254-AFF6-97E9B28DC867@isc.org> (Joe Abley's
message of "Tue, 24 Jan 2006 13:05:21 -0500")
Errors-To: owner-nanog@merit.edu
Joe Abley <jabley@isc.org> writes:
> On 24-Jan-2006, at 12:07, Robert E.Seastrom wrote:
>
>> He said "via two different autonomous domains", which I took to mean
>> two upstreams... and my understanding is that (on ciscos anyway)
>> you're talking per-packet, not per-flow load balancing.
>
> If you can get two candidate routes for the same destination into the
> FIB, then you'll get per-flow load balancing as long as CEF is
> running, no?
Yes and no. CEF is {src, dst} hash IIRC, and "per-flow" usually means
{src, srcport, dst, dstport, [proto, tos]} hash in my experience.
---Rob
