Path: utzoo!utgpu!news-server.csri.toronto.edu!bonnie.concordia.ca!uunet!mcsun!ukc!acorn!lsl!robin From: robin@lsl.co.uk (Robin Fairbairns) Newsgroups: comp.sys.dec Subject: Message-ID: <1991Jun28.084435.802@lsl.co.uk> Date: 28 Jun 91 08:44:35 GMT References: <81349@eerie.acsu.Buffalo.EDU> <23030002@acf3.NYU.EDU> Organization: Laser-Scan Ltd., Cambridge Lines: 20 In article <23030002@acf3.NYU.EDU>, tihor@acf3.NYU.EDU (Stephen Tihor) writes: > The DEQNA is likely to continue to work for some numbner of releases. After a fashion > However it will gradually become less and less effective as you upgrade > your network software and as your netowrk pushes the limits of the > hardware specs. I amnot sure who told you 5.3 was a magic release. > Its more a matter of DECNET-OSI (phase V) straining it to a level of > performance that some consider unacceptable and the like. [...] Performance of the DEQNA is already (and has been for a long time) considered unacceptable for NI cluster use. That's why NI clusters solemnly software checksum each packet if there's a DEQNA around. The things seem to have been designed before the typical volume of traffic on an ethernet reached its present levels. -- Robin Fairbairns, Senior Consultant, postmaster and general dogsbody Laser-Scan Ltd., Science Park, Milton Rd., Cambridge CB4 4FY, UK Email: robin@lsl.co.uk --or-- rf@cl.cam.ac.uk