| T.R | Title | User | Personal
Name | Date | Lines |
|---|
| 818.1 | | KONING::KONING | Paul Koning, A-13683 | Tue Dec 22 1992 10:27 | 7 |
| "Roving MAC" is a codeword for a complex non-standard feature that some
concentrator makers have added to their product, and attempt to sell as
something useful. There are lots of excuses for it, none technically
sound. For example, "testing" is something all FDDI devices do already
(using Link Confidence Test).
paul
|
| 818.2 | anymore data? | GRANMA::RHARPER | | Mon Jan 04 1993 11:55 | 4 |
| Do you know where I could get more info on it? I'd like to be able to
discuss it usefullness with my customer.
Rich
|
| 818.3 | In DEC products? | JUMP4::JOY | Happy at last | Mon Jan 04 1993 13:09 | 12 |
| I just saw a message from Bruce Sweet which says that the FDDI
Concentrator in the DEChub 900 with have a "Roving" MAC. I was shocked
to read this after all the bashing we've done around this topic for the
past several years. Is there really a good reason to have this in our
concentrator or are we just knuckling under to pressure from customers
who don't understand how it works?
First DAS PC adapters, now roving MACs.......we certainly have changed
our tune on a lot of issues lately.
Debbie
|
| 818.4 | No Roving MAC | QUIVER::PARISEAU | Luc Pariseau | Tue Jan 05 1993 08:30 | 15 |
|
In the HUB we will be able to place all PHYs and the MAC on the
Primary OR Secondary Path. There will be no Local Path.
It is an all or nothing deal. ALL Ports + MAC are always on the
same path together.
I guess it depends how you define Roving MAC, but I don't think
this is what people would expect. Our MAC will not be able to
move to a single port for 'testing' (as Paul mentioned).
rove : to wander about at random
I sure hope our MAC doesn't 'wander about at random'.
Luc
|
| 818.5 | Thanks for clarification | JUMP4::JOY | Happy at last | Tue Jan 05 1993 09:29 | 7 |
| Luc,
Thanks for the clarification. Our MAC sounds a bit more rational
than other vendors who use the term "roving". They seem to stick more
closely to the dictionary definition you gave.
Debbie
|
| 818.6 | | KONING::KONING | Paul Koning, A-13683 | Tue Jan 05 1993 10:09 | 4 |
| What was talked about is a dual path concentrator; that's quite another beast.
Such things are fully defined by the standard and have meaningful properties.
paul
|
| 818.7 | What is a Dual Path Concentrator? | DPDMAI::DAVIES | Mark, SCA Area Network Consultant | Tue Jan 05 1993 15:20 | 7 |
| OK. The DEChub 900 FDDI concentrator is a "dual path concentrator".
How is one of these different from a standard concentartor such as the
DC500?
Mark
|
| 818.8 | Empty secondary on DC500 | QUIVER::PARISEAU | Luc Pariseau | Wed Jan 06 1993 09:00 | 17 |
|
The old concentrator had a secondary path but the M ports and
MAC could only go on the primary path. The HUB concentrator
can place the M ports and MAC all on the primary PATH OR all on the
secondary PATH (be carefull..."primary ring" doesn't always mean
the same as "primary path"...I can explain that some other time).
For those that don't know why you would want to do this, the
idea is that now you can use the secondary ring AND the primary
ring simultaneously and get 200 Mbps IF your trunk ring is NOT
wrapped (which is normally true). Which also means you need a
bridge/router between your primary and secondary rings (like two
ports of a GIGAswitch...or just get a 22 port GIGAswitch to start
with and get 22 "rings"...22 X 100 = 2200 Mbps and a lot less
confusion!)
Luc
|
| 818.9 | path vs ring? | DPDMAI::DAVIES | Mark, SCA Area Network Consultant | Wed Jan 06 1993 09:19 | 11 |
| I am still confused. You say that either ALL the MACs/ports can placed
on the primary or secondary path, but not a few on each. This doesn't
seem to take advantage of the "200Mbps available" on both rings if all
MACs are on the same ring.
Maybe I need to understand better the relation of a path and a ring.
Regards,
Mark
|
| 818.10 | Perception = Reality | JUMP4::JOY | Happy at last | Wed Jan 06 1993 12:37 | 16 |
| Mark,
I'm taking a guess here, but I think the idea is that you would run
one set of fiber pairs to create your dual-ring, then you could place
multiple concentrators on either the primary or the seconardy "path"
with a bridge or router between them to create a simulated 200 Mbps
FDDI network. If you used a bridge, the fact that the systems connected
to the concentrators were on two seperate paths would be invisible
(given a full speed 100-100 bridge) so you could build something like a
cluster, with half the nodes on one path and half on the other and you
can give the impression of having 200 Mbps between nodes.
Maybe we're finally getting the hang of "marketing" our features like
SYnoptics and Cisco do....perception is reality, right?
Deb
|