home *** CD-ROM | disk | FTP | other *** search
- Newsgroups: comp.dcom.cell-relay
- Path: sparky!uunet!zaphod.mps.ohio-state.edu!uwm.edu!ux1.cso.uiuc.edu!usenet.ucs.indiana.edu!robelr@mythos.ucs.indiana.edu
- From: "Allen Robel" <robelr@mythos.ucs.indiana.edu>
- Subject: Re: Comment made at Next Generation Networks conference
- Message-ID: <By6GL7.EB0@usenet.ucs.indiana.edu>
- Sender: <robelr@mythos.ucs.indiana.edu>
- Organization: Indiana University
- Date: Mon, 23 Nov 1992 16:25:40 GMT
- Lines: 19
-
- Craig Partridge writes:
- > There doesn't have to be a conflict here. One can use FDDI as an access
- > protocol to an FDDI switch (just like switched Ethernet). So the question is,
- > how fast can one host drive the MAC protocol (FDDI or ATM equivalent or,
- > indeed, 100Mbit/s Ethernet)?
- >
-
-
- Hmm, does someone make an FDDI switch? If so, this certainly changes things.
- I'm a bit uncertain how efficient switching might be performed for FDDI
- in the same way that it is for ethernet. For instance, to switch an FDDI
- means, I would think, to go through the claim token and ring initialization
- processes each time that a frame is switched, right? Is there much overhead
- here? I'm not an FDDI expert so don't know 1) how much time this initialization
- process takes, and 2) if there might be a way around it. Ethernet, of course,
-
- doesn't have this problem so I agree with you about 100Mb/s enet switching...
-
- allen
-