Path: utzoo!attcan!uunet!lll-winken!lll-lcc!lll-tis!helios.ee.lbl.gov!pasteur!ucbvax!decwrl!labrea!rutgers!gatech!hubcap!Eugene From: eugene@pioneer.arc.nasa.gov (Eugene N. Miya) Newsgroups: comp.parallel Subject: Re: classifying/characterizing parallelism Message-ID: <1827@hubcap.UUCP> Date: 6 Jun 88 13:00:24 GMT Sender: fpst@hubcap.UUCP Lines: 28 Approved: parallel@hubcap.clemson.edu This is no longer algorithms..... >I think that the best way to think of pipelined parallelism under the Flynn >taxonomy is to call it MISD. In other words, the "unused" classification >among SISD, MISD, SIMD, MIMD does actually contain examples of real machines. The problem is MISD is it is difficult to define: you say it next: >I suppose the real difficulty is with definitions: What is an instruction? >What is data? What is a CPU? I've seen one group of of people call the C.mmp an MIMD computer and another call it a MISD. Pipelining? Why not? you can stretch, but you have just lost the characterization value. >Of course, assuming that the four Flynn classes are orthogonal is probably a >mistake. Just an added note: you said a CPU/ALU might be SISD, most are regarded as "bit-level" parallel (unless you have something bit-serial [AMD29*]. I sent Chris's note to Mike Flynn for comment. Another gross generalization from --eugene miya, NASA Ames Research Center, eugene@aurora.arc.nasa.gov resident cynic at the Rock of Ages Home for Retired Hackers: "Mailers?! HA!", "If my mail does not reach you, please accept my apology." {uunet,hplabs,ncar,ihnp4,decwrl,allegra,tektronix}!ames!aurora!eugene "Send mail, avoid follow-ups. If enough, I'll summarize."