[Scilab-users] * ParFor : a parallelized dedicated for loop control structure

[Samuel Gougeon]

http://mailinglists.scilab.org/Scilab-6-0-0-has-been-released-tp4035500p4035570.html

Le 22/02/2017 à 21:50, Amanda Osvaldo a écrit :
>
> ** ParFor*
>
> Forward parallelism, it would also be very convenient to be able to 
> parallelize operations within the macro without having to put the code 
> to be paralyzed in another macro.
>
> For example:
>
> data = [1:10];
>
> r = [];
>
> *parfor* i = 1:10
>
> r(i) = data(i) * 5; // Each operation will be performed in parallel
>
> endfor
>
> // You will only reach this line after all the parallel operations 
> have finished.
>
> disp(r);
>

Indeed, this would be a very handy and straightforward way to run a 
parallelized loop.

For the time being, although  its help page is still distributed,
https://help.scilab.org/docs/6.0.0/en_US/parallel_run.html
*/parallel_run/* has been removed from Scilab 6
--> parallel_run
Undefined variable: parallel_run

It was pointed to one year ago: http://bugzilla.scilab.org/14486

And some MPI features announced in the restricted 6.0.0 release notes 
are not actually distributed:
/* MPI_Create_comm create a new communicator from MPI_COMM_WORLD using 
MPI world ranks.//
/
--> MPI_Create_comm
Impossible to load mpi.dll library

at least on the MSWindows 64 bits plateform, for which the Scilab 
installer proposes by default to instal the MPI module.

So, parallelization is somewhat regressing. On windows, not really, 
since parallel_run never actually addressed more than one processor.

Samuel

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.scilab.org/pipermail/users/attachments/20170223/2f4b1371/attachment.htm>