Path: utzoo!attcan!utgpu!jarvis.csri.toronto.edu!mailrus!uwm.edu!uakari.primate.wisc.edu!aplcen!haven!umbc3!chimiak From: chimiak@umbc3.UMBC.EDU (Mr. William J. Chimiak ) Newsgroups: comp.dcom.lans Subject: Re: high speed networking between buildings Message-ID: <2333@umbc3.UMBC.EDU> Date: 27 Sep 89 12:15:25 GMT References: <4574@ursa-major.SPDCC.COM> <337@ai.etl.army.mil> <2314@umbc3.UMBC.EDU> <5971@b11.ingr.com> Reply-To: chimiak@umbc3.umbc.edu.UMBC.EDU (Mr. William J. Chimiak (MMA)) Organization: Univ of Maryland Baltimore County Lines: 20 > Broadband with FDDI? I don't follow. Someone enlighten me (nicely). > > tony goodloe I am sorry for perhaps being misleading, but it is my impression that the nature of the question is the nagging problem that workstations now have the I/O and processing speeds which blow past the capabilities of ETHERNET. Folks with image processing applications and huge file transfers are irritated with the LAN/CAN/WAN bottleneck. One of the major problems is the media access which roughly resides in the Data Link layer (layer 2) of the OSI model. FDDI is capable of delivering 100 Mbps at the MAC layer. Besides being that fast, FDDI has the potential of delivering real time performance - but this must be said with a caveat. Real time performance correctly implies an Application Layer capability. This is the final layer of the OSI model. To deliver true real time performance obviously implies that the layers three though six deliver the same real time performance. NIST is studyinging realtime performance of Layer 4 while folks like Gregg Chesson are developing the eXpress Transport Protocol (XTP) to at least give workstations the capability of operating on Gbps LANs.