Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Posting-Version: version B 2.10.2 9/5/84; site rpics.UUCP
Path: utzoo!watmath!clyde!burl!ulysses!allegra!mit-eddie!genrad!panda!talcott!harvard!seismo!rpics!schoff
From: schoff@rpics.UUCP (Martin Lee Schoffstall)
Newsgroups: net.dcom
Subject: dynamic t1 muxes
Message-ID: <203@rpics.UUCP>
Date: Mon, 28-Oct-85 18:39:20 EST
Article-I.D.: rpics.203
Posted: Mon Oct 28 18:39:20 1985
Date-Received: Wed, 30-Oct-85 06:31:26 EST
Distribution: net
Organization: RPI CS Department, Troy  NY
Lines: 32

I'm interested in some information on T1 muxes.  Specifically
I see the potential for having 3 T1 lines between points A and
B.  At site A I might have a PBX with 2 T1 interfaces to the
mux and a packet processor with 2 T1 interfaces to the mux.  Site
B is the same.  Now, it would be real easy to switch entire T1's,
during the "day" my PBX might have access to 2 of the T1's and
at night the packet processor might have them.  This seems to be
standard fare for T1 muxes and network management software.  What I would
like to have is more granuarity of sharing between the muxes,
at least at the level of DDS circuits (56kbits); BUT, I don't
want to deal with anything other than T1 interfaces to my
PBX or packet processor.  Effectively I might have during peak
hour on my packet network 2.0 X T1 bandwidth point to point and
"only" 1.00 x T1 bandwidth during idle hours, but say 1.33 x T1
during "normal" hours.  Does anyone know of anyone know of
something like this that is currently available.

Please, no conceptual criticisms of having multiple layers of
load balancing since this situation is a result of politics/geography.


-- 
marty schoffstall
schoff%rpics.csnet@csnet-relay	ARPA
schoff@rpics			CSNET
seismo!rpics!schoff		UUCP
martin_schoffstall@TROY.NY.USA.NA.EARTH.SOL	UNIVERSENET

RPI
Computer Science Department
Troy, NY  12180
(518) 271-2654