[12219] in Commercialization & Privatization of the Internet

home help back first fref pref prev next nref lref last post

ANS MCI IBM NSF

daemon@ATHENA.MIT.EDU (Joseph W. Stroup)
Tue May 10 02:34:31 1994

Date: Mon, 9 May 1994 21:10:59 -0700 (PDT)
From: "Joseph W. Stroup" <nettech@crl.com>
To: com-priv@psi.com
Cc: "Joseph W. Stroup" <nettech@crl.com>

I have done alot of research into the technical portion of the current 
backbone. I have talked to IBM engineers as well as many others connected 
with the MCI-IBM-ANS group. Other than the U.S. military I can find no 
T-3 network that would even come close in meeting the spec's set down by 
the NSF. Sprint does not have a T-3 network that can duplicate even the 
current backbone. Yes, they have announced ATM and T-3 with the Cisco 
7000 routers. The 7000 series does NOT pass data at 45 meg-a-bittys. It 
will however work at just a shade under 50 percent of that. Even ANS has 
not seen huge levels of traffic on their network and certainly nothing 
even close to T-3. I want to caution people. I am only talking about 
technical issues here nothing else. This is a very complex situation and 
the technical issues are only a part of the global picture. 

I do not think that Sprint could get a network up and running at true T-3 
speeds in any short order. I do not think that any other company would be 
able to do what the current group has done. At least not in any short 
amount of time. When it comes to 155 meg-a-bitty data IBM is in the 
unique position to use the information gathered from its past work. IBM 
should be able to construct the necessary hardware/software interface to 
allow an NT switch to operate at high speed over the MCI network. If they 
engineers are allowed the freedom to design without the management 
getting in their hair - We might all see a unique and creative approach 
to solving engineering problems and that would produce and excellent end 
product.

I have serious concerns over management, direction and motive when it 
comes to the NSF idea of the way to solve problems but, as far as the 
ability of IBM to make things work... I no longer think thats an issue.

For me its case closed. 

Joseph Stroup



home help back first fref pref prev next nref lref last post