Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP Path: utzoo!mnetor!seismo!uwvax!astroatc!johnw From: johnw@astroatc.UUCP (John F. Wardale) Newsgroups: comp.arch Subject: Re: chewing up mips with graphics Message-ID: <343@astroatc.UUCP> Date: Mon, 29-Jun-87 15:17:46 EDT Article-I.D.: astroatc.343 Posted: Mon Jun 29 15:17:46 1987 Date-Received: Wed, 1-Jul-87 02:12:39 EDT References: <8270@amdahl.amdahl.com> <359@rocky2.UUCP> Reply-To: johnw@astroatc.UUCP (John F. Wardale) Organization: Astronautics Technology Cntr, Madison, WI Lines: 40 In article <1323@ogcvax.UUCP> pase@ogcvax.UUCP (Douglas M. Pase) writes: > In article <2120@dg_rtp.UUCP> wood@dg_rtp.UUCP (Tom Wood) writes: > to make 100 compiles go 100 times faster by using 100 machines than it > In article (ME) johnw@astroatc.UUCP (John F. Wardale) writes: > > This and other related studys show that a large fraction of > problems that have no, or very limited parrallelism. > >As you have stated the study, it does *not* support your conclusion. Tight >loops in FORTRAN (shudder) programs may often be parallelized, by pipelining >instructions. Kuck's work on parallelizing compilers have shown amazing >improvements can be gained by pipelining and vectorizing DO loops. (Yes, >both are a form of parallelism.) It is the data dependencies which determine >the available parallelism, not the size of the code. Right, and *MOST* programs have dependencies that limit the paralleliam to one or two digits worth. Tom's original posting was looking for three digits worth. To get that, you need a problem that is "embarrassingly parallel" or an amazing system (hw+sw). >This seems pretty easy to me. Any > would benefit a lot from just about any level of parallelism. True, ***BUT*** Can you achive >= 100x speed-up by parallelizing the examples you give. Use as many processors as you want! You can have a case of from me if you succeed! John W - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - Name: John F. Wardale UUCP: ... {seismo | harvard | ihnp4} !uwvax!astroatc!johnw arpa: astroatc!johnw@rsch.wisc.edu snail: 5800 Cottage Gr. Rd. ;;; Madison WI 53716 audio: 608-221-9001 eXt 110 To err is human, to really foul up world news requires the net!