home *** CD-ROM | disk | FTP | other *** search
- Newsgroups: comp.dcom.cell-relay
- Path: sparky!uunet!mcsun!sunic!sics.se!craig
- From: craig@sics.se (Craig Partridge)
- Subject: re: Out of order messages
- Message-ID: <1992Aug28.160848.2723@sics.se>
- Organization: Swedish Institute of Computer Science, Kista
- Date: Fri, 28 Aug 1992 16:08:48 GMT
- Lines: 17
-
- > This is a simple question, but I can't seem to track down
- > a definitive answer: Is it possible for ATM cells when
- > providing a connection based service to become out of order
- > after traversing multiple switches? Some specifications
- > start from either answer in their development of protocols.
-
- I've yet to see an ATM spec that says cells can be re-ordered.
-
- I have seen a number of papers on cell network design that argue that
- permitting cell-reordering would be nice.
-
- The reason is that (at least with current state of the art),
- *ensuring* that no reordering occurs is rather expensive, and
- one can build cheaper switches (and more flexible LANs) if one
- does not have to ensure that no reordering occurs.
-
- Craig
-