[Beowulf] Has anyone actually seen/used a cell system?
diep at xs4all.nl
Mon Oct 2 09:18:58 PDT 2006
Actually Gofer also compiles to C.
That automatic translation had a cost a factor 200 in speed though.
There was some claim of factor 50 (but that was based upon a
very inefficient C compiler where it worked for, namely turbo c),
supported on paper. But that was old paper. I benched it at factor 200.
Even if you work hard and get it down to factor 50, still that's a factor 50
Cilk for example is just a few functions you can use from *inside* C code.
How is sequoia going to beat Cilk?
Just using a 'library' from within C/C++ is always better than ANL that
thereby losing big efficiency.
----- Original Message -----
From: "Andrew Shewmaker" <agshew at gmail.com>
To: "Vincent Diepeveen" <diep at xs4all.nl>
Cc: <J.A.Delcorso at larc.nasa.gov>; <beowulf at beowulf.org>
Sent: Monday, October 02, 2006 4:03 PM
Subject: Re: [Beowulf] Has anyone actually seen/used a cell system?
> On 10/2/06, Vincent Diepeveen <diep at xs4all.nl> wrote:
>> Not wanting to sound too negative, but total nonsense concept.
>> First of all this 'sequoia' claims to be a new programming language.
>> Meaning it'll take a year or 30 until some good compilers for it are
>> provided someone is going to support it.
>> Which isn't going to happen.
> Like many new programming systems, it compiles to C.
>> The parallellization basically is based upon complex assumptions for
>> programmers. So for programmers they don't actually make it easier than
>> trivial parallellization is via C/C++ function calls.
>> The sequoia parallellization basically is simplistically over for loops
>> a programmer himself can trivially parallellize too.
> Sequoia allows the same source to compile and run on systems with
> very different memory hierarchies. It uses MPI on clusters and DMA
> on the Cell. It also manages overlays on the Cell. Do you consider a
> portable runtime system that manages overlays and streams data
> asynchronously trivial to implement?
>> Further the optimization of sequoia simply doesn't happen. They assume
>> "kernel libraries" solve the problem. Interestingly it mentions
>> "if kernel libraries could be obtained, such as FFTW and the intel MKL
>> PCs, or the IBM SPE matrix library for Cell, we call these libraries from
>> Sequoia leaf tasks".
>> In short if some algorithm has not been implemented for sequoia, sequoia
>> unusable. Others may do the work as usual to promote sequoia.
> As I understand it, the leaf tasks can be written in C, Fortran, or
> Saying Sequoia is unusable is like saying that MPI is unusable.
> Andrew Shewmaker
More information about the Beowulf