Path: utzoo!utgpu!jarvis.csri.toronto.edu!rutgers!att!cbnewsh!tds
From: tds@cbnewsh.ATT.COM (antonio.desimone)
Newsgroups: comp.protocols.tcp-ip
Subject: Re: Traffic Sensitive SPF Routing is NOT too hard!
Message-ID: <6203@cbnewsh.ATT.COM>
Date: 1 Dec 89 05:11:53 GMT
References: <8911302038.AA19457@ucbvax.Berkeley.EDU>
Organization: AT&T Bell Laboratories
Lines: 33

From article <8911302038.AA19457@ucbvax.Berkeley.EDU>, by jzinky@BBN.COM ("John A. Zinky"):
> I would like to give my opinion on traffic sensitive routing from the
> experience of running large over-subscribed networks, such as the 1987-88
> ARPANET.

Or the phone network?  There is in fact a lot of experience in running
large networks and using traffic-sensitive routing.  When I started
this thread (I think we're on the same thread) I was really interested
in how dynamic routing algorithms in datagram networks compare to the
algorithms used in circuit-switched networks, to see if my intuition
is completely useless for datagram networks.  An apparent difference
is that routing in datagram networks can (in principle) react to
congestion on the timescales on which queues build up.

> equipment procurement (months to years). Routing is an allocation
> policy that maps traffic flows onto available resources. It works on
> the time scale of network operations (minutes to days).  Congestion
> control regulates user traffic so that resources are not
> oversubscribed. It works on the time scale of a few round trip times.

This discussion gives me a nice warm feeling since it says that routing
in datagram networks, in practice, behaves much like routing in
circuit-switched networks in the sense of mapping traffic flows, *not*
individual packets.

> "Be careful, it's a real world out there"

I love this guy!
-- 
Tony DeSimone
AT&T Bell Laboratories
Holmdel, NJ 07733
att!tds386e!tds


Brought to you by Super Global Mega Corp .com