| Title: | FDDI - The Next Generation |
| Moderator: | NETCAD::STEFANI |
| Created: | Thu Apr 27 1989 |
| Last Modified: | Thu Jun 05 1997 |
| Last Successful Update: | Fri Jun 06 1997 |
| Number of topics: | 2259 |
| Total number of notes: | 8590 |
Dave Katz Merit/NSFNET write in his paper "The Use of Connectionless Network Layer Protocols over FDDI Network" ...The main trunk of an FDDI network consists of dual counterrotating rings, referred to as the Primary and Secondary rings. In normal operations, the two rings form independent token paths. Each ring can be used for the transmission of data. If a component fails that breaks the transmission path in one of the rings, the nodes on either side of the failure splice the two rings together, forming one folded ring.... ... A MAC layer allows a station to send and receive data. Stations with only a single MAC may choose to attach the MAC to either of the rings. Can someone tell me if the FDDI standard allows a SAS on the primary or secondary ring, so that I would get 2 x 100 Mbps bandwith (although no connectivity) ? I know that our implementation will not allow that ? SAS is attached to DAC. Cheers Walter.
| T.R | Title | User | Personal Name | Date | Lines |
|---|---|---|---|---|---|
| 207.1 | May be it could be done, but... | CGOO01::HARROP | Ring those phones!!! | Thu Feb 21 1991 14:59 | 13 |
I was in PID (I am in sales so I was listening, not presenting) where
the customer asked this same question. The answer had two parts to
it:
1) that's not how FDDI is implemented, but it might be able to be done.
It is definitely non-standard, not part of the FDDI standard, and not
supported by Digital. If the customer really wanted to pursue this on
their own, Digital would not stop them, but Digital had no interest in
it due to lack of adherence to standards.
2) implementing such a thing would eliminate the redundancy built into
FDDI, another reason that Digital would not be interested.
| |||||
| 207.2 | That's mostly incorrect! | KONING::KONING | Lietuva laisva! | Tue Feb 26 1991 12:30 | 12 |
Who gave that answer? Part (1) is false. Part (2) is basically correct: you could get twice the bandwidth but only when there are no faults. When a fault occurs and the ring wraps, your bandwidth is cut in half. If you want redundancy, you should run multiple LANs. That gives you much better control over what happens when faults occur. It also lets you have triple or higher redundancy; the "secondary ring" approach allows only one redundant connection and therefore has very limited flexibility. paul | |||||