| 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 |
We have a customer who is planning an FDDI based metropolitan area
network which will bridge Ethernet (and Token ring) networks. They
have asked us about the timing tolerences for differenet Ethernet based
protocols/services including: LAD, LAST, DECnet and LAT. I realize
this sounds like an application dependent question but is there
anything the FDDI folks can say about the use of these different
services on high speed extended LANs. I have posted this question in
the PCSAV2 notes file also.
I think what the customer would like is to be able to state throughput
timing goals (in milliseconds or something) for the network and then
commit to maintaining the average service levels. They want to make
sure their goals are consistent with the needs of common services like
DECnet/PCSA.
| T.R | Title | User | Personal Name | Date | Lines |
|---|---|---|---|---|---|
| 78.1 | KONING::KONING | NI1D @FN42eq | Mon Jun 25 1990 17:18 | 12 | |
FDDI allows a ring circumference of up to 200 km. At that sort of distance the channel delay is still under 2 ms. None of the protocols you mentioned should suffer unduly. For comparison, the "7 bridge rule" is based on an assumption of 1 ms "typical, loaded" delay through a bridge. In other words, such an oversized FDDI could be counted as if it were 1 or 2 bridges, and the 7 bridge rule applied as usual. On the other hand, we have configuration recommendations for FDDI, and MAN sized FDDIs are well outside those recommendations. paul | |||||
| 78.2 | arithmetics | NCEIS1::CHEVAUX | Patrick Chevaux, Nice, 828-6995 | Wed Oct 10 1990 11:23 | 18 |
Back to basic arithmetics:
- propagation in the fiber at .66c (example):
1ns = 20cm
1us = 200m
1ms = 200km <--- max circumference of FDDI ring
1 s = 200000km
- send/receive delays
at 10 Mbps 1500Bytes=12000bits takes 1.2ms
at 100Mbps _________ _________ _____ 120ns
- I assume the bridges complete the receive operation before they
forward the frame.
- I haven't included the token acquisition time
| |||||
| 78.3 | ! | ZPOVC::HWCHOY | It must be Thursday. | Wed Oct 10 1990 12:59 | 7 |
� - propagation in the fiber at .66c (example):
but propagation in the fiber should be "c"!?
ain't it?
| |||||
| 78.4 | Glass is slower than a vacuum | BAGELS::LEVY | Wed Oct 10 1990 13:27 | 5 | |
re: -.1
Nope. Propagation in a vacuum is "c"
Propagation in a fiber (n~1.5) is ~.66c
| |||||