[GRASS-dev] Interested in parallelization of GRASS

Maris Nartiss maris.gis at gmail.com
Sat Mar 26 05:37:46 EDT 2011


Hello,
as nobody has answered Your mail, I would drop my 0.02$.

As most of current GRASS is designed with single process single thread
short-lived CLI applications in mind. Redesigning core of GRASS to
take advantage of modern multi-core CPUs and/or nodes /GPUs would be a
nice thing. You can read about current state of GRASS parallelization
at WIKI [1].

Keep in mind that GRASS mostly is written in C and there are no plans
to change core language to anything else.

There are many areas that could benefit from thread-safe/parallel
processing: vector and raster reading/writing, OGSF etc.

As I'm not a (good) programmer, I can't help You more. Feel free to
explore existing codebase [2] and documentation [3], [4].


Maris.

1. http://grass.osgeo.org/wiki/Parallel_GRASS_jobs
2. http://trac.osgeo.org/grass/browser/grass/trunk
3. http://grass.osgeo.org/programming7/
4. http://trac.osgeo.org/grass/


2011/3/21, Vivek Yadav <vicky.it.bhu at gmail.com>:
> Hello,
>                    I would like to work on parallelization of GRASS
> libraries through implementation of OpenMP in thread-safe algorithms and
> modification of non-thread-safe algorithms to make them thread-safe if
> possible.
> I have good experience in C++( Templates/STL, generic programming,
> tag-dispatching and basic idea of meta-programming). Also, I have worked
> with OpenMP to parallelize time-averaged simulation.
>                   I would like to know more about this project and apply
> under the GSoC program.
>
> With Regards,
>                 Vivek.
>


More information about the grass-dev mailing list