Path: utzoo!attcan!utgpu!jarvis.csri.toronto.edu!mailrus!ncar!gatech!hubcap!Eugene From: eugene@eos.arc.nasa.gov (Eugene Miya) Newsgroups: comp.parallel Subject: Re: Class 7 definition Message-ID: <6580@hubcap.clemson.edu> Date: 25 Sep 89 18:59:25 GMT Sender: fpst@hubcap.clemson.edu Lines: 27 Approved: parallel@hubcap.clemson.edu In short: This is none. Any person who will quote you sizes or speeds: Robins Williams recently said: If you can remember Woodstock, you weren't there. Numerical classification is an artifact of what is now the Dept. of Energy. The person you want to speak to who developed it is Sid Fernbach, formerly at LLL now a consultant to CDC. You had best speak soon as Sid is having health problems. He lives in San Jose. Classes are a numeric categorization of current fastest supercomputer. Hence, if a machine isn't a supercomputer (sic) then it doesn't rate. E.g., a VAX is many times more "powerful" [what ever that measurement is] than say many early computers, but there is no temporal basis for comparison. Hence the fastest class 6 machines were Crays, CDCs (earlier), NEC, Fujitsu, Hitachi. None of the "mini-supers" rated on this scale. "Classes" stopped growing about 1984 when the Cray-2 was released. I have pressed and pressed Sid for a definition of class 7 and whether examples for such exist. It's not clear. Further, machines like the connection machine and hypercube complicate the issue. Any person who can give simple definitions to what a Class 7 machine is only fooling himself and does not understand the issues....... yet. --eugene