Path: utzoo!news-server.csri.toronto.edu!cs.utexas.edu!swrinde!elroy.jpl.nasa.gov!lll-winken!llnl!physics.llnl.gov!brooks From: brooks@physics.llnl.gov (Eugene D. Brooks III) Newsgroups: comp.arch Subject: Re: massive parallelism, was CDC 6600 and TI ASC Message-ID: <780@llnl.LLNL.GOV> Date: 11 Mar 91 01:58:57 GMT References: <45252@ut-emx.uucp> <1991Mar7.215545.430@zoo.toronto.edu> Sender: news@llnl.LLNL.GOV Organization: Lawrence Livermore National Laboratory Lines: 20 In article lamson@el1.crd.ge.com (scott h lamson) writes: >Given this line of reasoning, how do you look at massive parallel ala >the connection machine? Should you think of the CM as a slow scalar >machine with a super fast very long vector processor? Yes, this is the correct view of the CM-2. >If so, will it fail for the same reasons you give? Yes, it will fail for the same historical reasons. NASA AMES has some impressive computational efficiencies for this machine on 3-D FFTs, the number I saw on a viewgraph was "~0" >One difference is vector units >take only 1 clock cycle for each additional operation, whereas CM's >take no extra clock cycles for each additional operation (until you >run out of processors). This does not matter. >but this has nothing to do with the scalar >speed argument anyway. Right. What gets you is vector operations under mask. The fundamental capability which is needed is MIMD functioning of the processors, so that each processor can go its own way and get effective work done instead of being disabled for the computation.