home *** CD-ROM | disk | FTP | other *** search
- Newsgroups: comp.dcom.cell-relay
- Path: sparky!uunet!panther!mothost!merlin.dev.cdx.mot.com!pjd.dev.cdx.mot.com!peterd
- From: peterd@pjd.dev.cdx.mot.com (Peter Desnoyers)
- Subject: Re: SSCOP better than TCP?
- Message-ID: <peterd.721416994@pjd.dev.cdx.mot.com>
- Sender: news@merlin.dev.cdx.mot.com (USENET News System)
- Nntp-Posting-Host: pjd.dev.cdx.mot.com
- Organization: Motorola Codex, Canton, Massachusetts
- References: <1992Nov9.181708.24894@sics.se>
- Date: Tue, 10 Nov 1992 17:36:34 GMT
- Lines: 34
-
- craig@sics.se (Craig Partridge) writes:
-
- >goldstein@carafe.enet.dec.com (Fred R. Goldstein) writes:
-
- >> Just in case anyone's wondering, SSCOP (service-specific
- >> connection-oriented protocol) is the service-specific sublayer that sits
- >> above AAL3/4 OR AAL 5 to create a connection-oriented data link
- >> service.
- >> ...
- >> Or if you want "end-to-end ATM", it can be used to deliver
- >> a transport-like service.
- >
- >Sigh... I thought everyone had learned about the end-to-end argument by
- >now. The link layer (ATM) has no business providing a function that higher
- >layers also have to provide. If this synopsis is accurate, SSCOP sounds like
- >a remarkably dumb idea.
-
- Actually, some of Fred's earlier postings lead me to believe that he
- thinks SSCOP is a good idea *because* of the end-to-end argument. (sorry
- if I'm mis-representing you, Fred...)
-
- The end-to-end argument doesn't say "never re-transmit at the link layer";
- it says "retransmit at the link layer for efficiency, not reliability".
-
- The argument for SSCOP would then be that (1) ATM service will lose
- many cells under congestion, and (2) SSCOP would be more efficient
- than e.g. TCP at retransmission.
-
- I'm curious about the second point - I don't see why any protocol above
- the AAL layer (with the same window strategy) would be any better than
- another at dealing with cell loss.
-
- Peter Desnoyers
- --
-