home *** CD-ROM | disk | FTP | other *** search
- Path: sparky!uunet!cis.ohio-state.edu!zaphod.mps.ohio-state.edu!cs.utexas.edu!usc!isi.edu!finn
- From: finn@isi.edu (Greg Finn)
- Newsgroups: comp.dcom.cell-relay
- Subject: Re: How does one compare tansport systems ?
- Message-ID: <22931@venera.isi.edu>
- Date: 20 Nov 92 22:44:37 GMT
- References: <1992Nov18.025754.14749@trl.oz.au>
- Sender: news@isi.edu
- Reply-To: finn@dalek.isi.edu (Greg Finn)
- Organization: USC-Information Sciences Institute
- Lines: 33
-
- In article <13566@grayt> grayt@Software.Mitel.COM (Tom Gray) writes:
-
- >Traffic goes application to application ie user to user ie end to end.
- >ATM addresses the building of user applications which are beyond the
- >traditional data applications now running on LAN's.
-
- Name one. ATM is not a transport layer protocol. It isn't
- end-to-end. With the addition of call set-up protocol, an AAL layer,
- and an as yet undefined multicast layer, each residing above it, ATM
- may be thought of as a part of a transport layer.
-
- >It does this while retaining the capability of running the traditional data
- >applications. Indeed it does this while reducing the complexity of
- >a local network making it more manageable and scalable.
-
- You are labouring under some misunderstandings. At the very
- least ATM requires fragmentation and reassembly in host interfaces.
- That is not required for data-link layers that support variable-length
- packets. A four-in/four-out 500 Mb/s per-channel variable-length
- router and two DMA drivers already fit onto < 10% of a microprocessor
- chip's logic area. Designed and developed by Caltech, we use them,
- the floor below me has them fabricated, proof by silicon. We don't
- need call setup. We don't need fragmentation or reassembly. We don't
- need AAL. We support TCP/IP.
-
- ATM will not succeed in the LAN domain because of technical
- superiority. It may succeed because of marketing superiority or
- decrees by government. That happens a lot. Marketing and politics
- are things that I know nothing about. I also have no informed opinion
- about what is well suited to the all-optical networking domain.
- --
- Gregory Finn (310) 822-1511
- Information Sciences Institute, Marina Del Rey, CA 90292
-