Tony Li <tli@cisco.com> (05/06/91)
Luis,
I can't imagine that the MILNET is more complex then the rest of
the INTERNET but I haven't heard of any major problems on NFS-NET.
That depends on your metric. Currently NSF is distributing about 1800
routes and the Milnet cores are handling approximately 2300.
Can someone please explain how routing is handled on NFS-NET for
example??
Each of the border gateways on the NSF T3 backbone speaks EGP or BGP
to the regional nets. Exterior information is carried within the
backbone via BGP. Interior routing is done with a variant of IS-IS.
[This information is second-hand. Please pardon me if I've garbled
it.]
Tonystjohns@umd5.umd.edu (Mike St. Johns) (05/07/91)
On the subject of complexity in the MILNET... The MILNET consists of approximately 230 BBN C30/C300 packet switching nodes generally connected by 56KBit trunks. When I left DDN about 18 months ago, the average number of trunks per node was somewhere between 3 and 4 - in other words, somewhere between 345 and 460 trunk circuits. Contrast this with about 25 nodes and 40 trunk circuits for the NSF backbone. Later, Mike