Path: utzoo!utgpu!jarvis.csri.toronto.edu!mailrus!uwm.edu!gem.mps.ohio-state.edu!brutus.cs.uiuc.edu!husc6!bunny!hhd0 From: hhd0@GTE.COM (Horace Dediu) Newsgroups: comp.arch Subject: Re: parallel systems Message-ID: <7651@bunny.GTE.COM> Date: 17 Oct 89 19:02:16 GMT References: <35825@lll-winken.LLNL.GOV> <20336@princeton.Princeton.EDU> Organization: GTE Laboratories, Inc., Waltham, MA Lines: 80 In article <20336@princeton.Princeton.EDU>, mg@notecnirp.Princeton.EDU (Michael Golan) writes: > This came for various people - the references are so confusing I removed them > so as not to put the wrong words in someone's mouth: > > >>>Supercomputers of the future will be scalable multiprocessors made of many > >>>hundreds to thousands of commodity microprocessors. > >> > >This is the stuff of research papers right now, and rapid progress is being > >made in this area. The key issue is not having the components which establish > >the interconnect cost much more than the micros, their off chip caches, > >I currently lean to scalable coherent cache systems which minimize programmer > >effort. The exact protocols and hardware implementation which work best > >for real applications is a current research topic. > > 1) There is no parallel machine currently the works faster than non-parallel > machines for the same price. The "fastest" machines are also non-parallel - > these are vector processors. > Consider the 8k processor NCUBE 2--"The World's Fastest Computer." (yes, one of those). According to their literature: "8,192 64 bit processors each equivalent to one VAX 780. It delivers 60 billion instructions per second, 27 billion scalar FLOPS, exceeding the performance of any other currently available or recently announced supercomputer." It's distributed memory .5MB per processor, runs UNIX, and is a hypercube. I don't know the price, but I bet it's less than a Cray. Interesting to talk about GigaFLOPS. This is fast. > 2) A lot of research is going on - and went on for over 10 years now. As far > as I know, no *really* scalable parallel architecture with shared memory exists > that will scale far above 10 processors (i.e. 100). And it does not seems to > me this will be possible in the near future. Who cares about shared memory? Distributed is the only way to scale. Everybody realizes this since it can be proven. The only reason shared memory machines exist is because we don't yet know how to make good distributed machines. (Yeah, right! tell that to Ncube) IMHO shared memory is a hack using available bus technology while waiting for the real parallel machines to come. (they're already here) > 3) personally I feel parallel computing has no real future as the single cpu > gets a 2-4 folds performance boost every few years, and parallel machines > constructions just can't keep up with that. It seems to me that for at least > the next 10 years, non-parallel machines will still give the best performance > and the best performance/cost. This is very ambiguous. Parallel machines can use off-the shelf CPU's. If a fast micro is available then you can design a parallel machine around it as you would any workstation. The other problem: if cpu's increase 2-4 folds every few years and if this can be maintained for 10 years you can only expect a 32 fold increase. This is nothing. You can't expect problems to stay that small. If you expect to go beyond that you'll hit a wall with the fundamental boundary of the speed of light. You can't drive clock rates to infinity. The only way to speed up is to do it in parallel. Sure it's hard to program, but it's a new field, the tools are rudimentary and only hardware people are involved in their development. If enough effort is put into it parallel machines should not be any harder to program than your basic workstation. > time that really matters. And while computers get faster, its seems software > complexity and the need for faster and faster machines is growing even more > rapidly. Of course. To solve hard problems you *need* parallel execution. It's no secret that every big iron maker and every supercomputer shop is developing parallel machines. These are still modest efforts (<100 cpu's), but the leading egde is now in the 10k coarse grained, 64k fine grained processors. This should scale nicely to 1M processors in the next decade. After that we can expect some kind of new barriers to come up. > Michael Golan > mg@princeton.edu -- Horace Dediu \"That's the nature of research--you don't know |GTE Laboratories (617) 466-4111\ what in hell you're doing." `Doc' Edgerton |40 Sylvan Road UUCP: ...!harvard!bunny!hhd0................................|Waltham, MA 02254 Internet: hhd0@gte.com or hhd0%gte.com@relay.cs.net..........|U. S. A.