How Many Processor Cores Are Enough?
|
From: ranjit_mathews@yahoo.com on 12 Oct 2006 10:41 Nick Maclaren wrote: > In article <1160660486.912240.193460(a)k70g2000cwa.googlegroups.com>, > "ranjit_mathews(a)yahoo.com" <ranjit_mathews(a)yahoo.com> writes: > |> * At the time the Cray XMP was released, would you have been able to > |> dream of many of the current uses for PCs surpassing it in computing > |> capacity? > > Yes. I did. And, more interesting, there is documentary evidence that > some people older and more far-sighted than I am did so in the 1950s. Very perspicacious! What potential do you forsee for a Software IC implemented on a grid of processors and usable from serial codes (speedups from parallelism would last only for the duration of each call to the Software IC)?
From: Nick Maclaren on 12 Oct 2006 11:16 In article <1160664104.209124.194720(a)b28g2000cwb.googlegroups.com>, "ranjit_mathews(a)yahoo.com" <ranjit_mathews(a)yahoo.com> writes: |> |> Very perspicacious! What potential do you forsee for a Software IC |> implemented on a grid of processors and usable from serial codes |> (speedups from parallelism would last only for the duration of each |> call to the Software IC)? What on earth is a Software Integrated Circuit? Regards, Nick Maclaren.
From: Eugene Miya on 12 Oct 2006 11:58 In article <1160660486.912240.193460(a)k70g2000cwa.googlegroups.com>, ranjit_mathews(a)yahoo.com <ranjit_mathews(a)yahoo.com> wrote: >Nick Maclaren wrote: >> In article <1160646792.020194.135340(a)i42g2000cwa.googlegroups.com>, >> "ranjit_mathews(a)yahoo.com" <ranjit_mathews(a)yahoo.com> writes: >> |> Combinatorial problems >> |> Take, for example, planning a route for your vacation, >> |> using a super-sophisticated successor of Microsoft Mappoint. >> >> It has always been trivial to soak up an arbitrary amount of computer >> time. So what? > >So, as long as there are a few popular applications that do nifty stuff >quickly using many processors, not all applications would have to be >heavily parallelized for people to go for many processors. >> Ifr you look at those problems in more depth, you will find that an >> infinitesimal number of them justify the use of a full search; in >> practice, all that is needed is a near-optimal solution. That is >> why many 'insoluble' problems have been solved commercially for over >> 4 decades. > >You can get a near-optimal solution from Mapquest. But then, you can't Bad example. Mapquest: all kinds of errors. >tell Mapquest you like, say, lightly traveled 2 lane country roads. ... >and this is but one example, there might be many other applications >that developers have barely started dreaming* of, that can put many >processors to good use. >* At the time the Cray XMP was released, would you have been able to >dream of many of the current uses for PCs surpassing it in computing >capacity? 20 year old architecture. About 25% of the people who had Crays at that time had the foresight that micros would surpass Cray at the time. But then Cray had ideas which are faster than PCs today. The people who acquired X-MPs know how to utilize them. The rest of the world doesn't. This is why they get the bucks. Search for the joke about requiring Cray-class PC-compatibility in the arch news group of the time. --
From: ranjit_mathews@yahoo.com on 12 Oct 2006 14:28 Nick Maclaren wrote: > In article <1160664104.209124.194720(a)b28g2000cwb.googlegroups.com>, > "ranjit_mathews(a)yahoo.com" <ranjit_mathews(a)yahoo.com> writes: > |> > |> Very perspicacious! What potential do you forsee for a Software IC > |> implemented on a grid of processors and usable from serial codes > |> (speedups from parallelism would last only for the duration of each > |> call to the Software IC)? > > What on earth is a Software Integrated Circuit? It varied from one touter to the next. Think in terms of a design tool that can emit logic for either an FPGA or for a grid of processors. (Perhaps the tool would come with a runtime system to implement a dataflow machine on a grid of processors and emit logic that invokes the runtime system)
From: kenney on 12 Oct 2006 14:31
In article <egl6l3$69q$1$8300dec7(a)news.demon.co.uk>, tfb(a)tfeb.org (Tim Bradshaw) wrote: > I'm sure Dennis meant the speed of light in vacuo: I certainly > did. It is not something that should be used for calculating delays on chips. That depends on a lot of factors. In all cases it will be considerably less than C. Ken Young |