[SoC] Interested in GPU-based processing

Wolf Bergenheim wolf+grass at bergenheim.net
Thu Apr 1 14:46:08 EDT 2010


On Thu, Apr 1, 2010 at 17:59, Seth Price <seth at pricepages.org> wrote:
> Thanks for the replies all! Looking at the wiki page, it looks like r.proj
> and v.proj are already there. I'd also be interested in working on r.sun
> (it's related to my thesis). However, I'm concerned that there isn't enough
> time in a SoC project to translate many modules.
>
> How long will it take to alter the build process to detect an OpenCL
> install?

Seth,

We can make a branch for you to work on, since you might be touching
core parts. That way you can do it yourself and get to working on it,
without having to wait. During the SoC we can then integrate parts
back, when they are more complete.

GRASS uses autoconf to detect things, so it is probably not that hard to do it.

+1 on OpenCL for me too since I don't have an nVidia card :P

--Wolf


> ~Seth
>
>
> On Apr 1, 2010, at 12:27 AM, Markus Neteler wrote:
>
>> On Thu, Apr 1, 2010 at 12:45 AM, Seth Price <seth at pricepages.org> wrote:
>>>
>>> Hey all, I'm interested in writing GPU-based code to improve the speed of
>>> GRASS and GDAL as a GSoC project. It's such a basic operation that is
>>> used
>>> often, I think it'll be very useful for a broad audience. (I know it will
>>> in
>>> my own work.) I have a few questions for the community:
>>> 1) Would both GRASS and GDAL be interested in this project? It's my
>>> intent
>>> to write the code for both projects, but I'd like to know if there's a
>>> lack
>>> of interest.
>>
>> On the contrary, we are very interested to get GPU support into GRASS.
>> Some discussions have already taken place:
>>
>> http://grass.osgeo.org/wiki/GPU
>>
>> See links therein for discussion about CUDA versus OpenCL. Please
>> don't hesitate to bring this up on the grass-dev mailing list since some
>> experts are not reading here to my knowledge.
>>
>> ...
>>
>>> 4) What other GRASS modules are most in need of a speed boost? My current
>>> plan is only to improve resampling and reprojecting operations over the
>>> summer, but I'm curious what else I should look at.
>>
>> A selection could be also driven by popularity of a module, say, some
>> commands
>> are way more often used than others. I started a list in above mentioned
>> Wiki
>> page which I'll send to grass-dev for discussion.
>>
>> thanks for your interest,
>> Markus
>>
>> --
>> http://grass.osgeo.org
>> http://gis.fem-environment.eu
>
> _______________________________________________
> SoC mailing list
> SoC at lists.osgeo.org
> http://lists.osgeo.org/mailman/listinfo/soc
>



-- 
<8  )---- Wolf Bergenheim ----(  3>


More information about the SoC mailing list