[scilab-Users] Scilab use with multiple core processors

[michael.baudin at contrib.scilab.org]

 Hi,

 Yes and no : it depends on what you exactly mean by parallel
 computing.

 Yes, Scilab can use multicores.
 By default, Scilab v5 can use all your processors on Windows, for
 some operations.
 We can do the same on Linux, with a little more work (see the
 document below for details).

 For dense linear algebra, the library which makes this possible
 is the Intel MKL on Windows and the ATLAS on Linux.
 For Fourier Transform, this is done by FFTW.

 You can find more details in "Programming in Scilab",
 section 5 "Performance" :

 http://forge.scilab.org/index.php/p/docprogscilab/downloads/

 The core of the section is vectorization.
 Actually, there is more than multicore : even on a single
 core, the Intel MKL or ATLAS libraries make a very efficient
 use of the processor.

 You can also do GPGPU computing with the module by
 Delamarre :

 http://atoms.scilab.org/toolboxes/sciGPGPU

 provided that your video card support Cuda and doubles.

 This is very different from distributed computing.
 This can be done with the MPI branch of Scilab :

 http://gitweb.scilab.org/?p=scilab.git;a=shortlog;h=refs/heads/MPI

 As far as I know, the MPI project for Scilab is still in
 experimental stage, but you may ask Ledru for details.

 Best regards,

 Michaël


 On Mon, 25 Jun 2012 08:38:58 +0800, Grant Ellis <grant.ellis at gmail.com> 
 wrote:
> Hello,
>  
> Does Scilab allow use of multiple cores for parallel processing
> applications?  If not, is this planned for a future release?
>  
> Grant