[1500] in North American Network Operators' Group
Re: value of co-location
daemon@ATHENA.MIT.EDU (Sean Doran)
Tue Jan 23 11:41:00 1996
From: Sean Doran <smd@icp.net>
To: curtis@ans.net
Cc: nanog@merit.edu
Date: Tue, 23 Jan 1996 11:15:33 -0000
| This is all very amusing.
And pretty accurate, actually. The very best of Dennis. :)
| If ATM is being used in VBR without the V mode, essentially providing
| point to point connections between routers at above DS3 rate, then
| there is no need for complex reassembly or any form of congestion
| control. That may turn out to be the way ATM is used by ISPs.
Ok, so in order to do VBR without the V mode, one effectively
has to either *really* trust one's ATM network provider not
to have any congestion which could lead to cell loss as we
have seen pretty clearly that even very minimal, occasional
cell loss is deadly -- or one has to do the UUPSI hack:
purchase clear-channel L0 circuits and run ATM as a L1
protocol on top of it.
That is, the only ATM switches are the ones at the end points
and owned by the NSP.
Now, given that at above-DS3 speeds, an L0 circuit is almost
certainly to be provisioned using SONET, one could reasonably
prefer to run PPP over OC3c rather than ATM over OC3c.
This is much the same as running PPP over DS3 rather than
ATM over DS3, or PPP over DS1 rather than Frame Relay over DS1.
It is my prediction that by this spring, the only reasons why
anybody would want to run ATM for connecting pieces of the
Internet would involve unavailability of SONET/SDH from one's
carrier(s), the desire to save money on point-to-point
circuits by letting one's carrier(s) sell one VCs over some
unknown fabric rather than real circuits over their
transmissions infrastructure, an already-installed ATM
fabric, or general insanity.
(Well, general insanity may be the case with the UUPSI hack,
however both may well be driven much more by the cheaper port
density of Cascade FR switches, although they never seem to
admit to this in public; this could come into play with ATM
as well...)
| The
| other way will be as a LAN technology where the SAR complexity and
| congestion issues will come into play.
Assuming a reasonably inexpensive PPP-over-SONET/SDH
implementation that can do SONET/SDH back-to-back, and
a router which can move traffic among enough 155Mbps
point-to-point connections, it may well be more economical
to build a PPP-over-SONET/SDH mesh among one's routers rather
than play with ATM.
Of course, if for whatever reason one is running ATM on long
haul links, and has invested money in a multiport ATM
switch, one may well (for money reasons) want to jump
headlong into complete insanity by playing with the
complexities Curtis mentions.
Finally, as is the case in all the growth to date, the
better technology and design will likely reveal itself when
both methods saturate, and as bandwidth requirement scales.
Sean.
