[gdal-dev] CUDA PyCUDA and GDAL

Frank Warmerdam warmerdam at pobox.com
Wed Nov 18 20:19:52 EST 2009

Shaun Kolomeitz wrote:
> Thanks Seth,
> It makes sense that the slowest part of the whole equation would be the
> disk operations, and there must be quite a number of disk reads/writes
> when processing imagery. Currently we use Raid Arrays that push data
> through at a rate of 300MB/s, granted if these were SSDs in Raid0 we
> could push beyond 1GB/s. Currently to process (mosaic) an 80GB image it
> takes several days to complete. This is also on 32bit hardware, and I
> suspect is only single threaded so we're limited to 3GB RAM. From what I
> understood the most optimal caching size in GDAL is 500MB, using a 512M
> window (unless that has changed).
> If you can easily lay your hands on your GSoC application than that
> would be great. We are discussing what might be possible with a very
> talented coder, who eats these types of "challenges" for breakfast !
> Perhaps a better approach would be to use something like a grid
> computing approach then like Condor to break up the processing ?


I suspect that there is a gross issue with how the warping is being done,
and that it could be sped up substantially. On my two year old consumer
grade desktop I'm able to warp/reproject around 30GB/hour with gdalwarp.

I don't want to dissuade you from investigating CUDA options, but
you might be able to get ten fold improvement by hiring an experienced
gdalwarp developer for a few hours to investigate your particular

Best regards,
I set the clouds in motion - turn up   | Frank Warmerdam, warmerdam at pobox.com
light and sound - activate the windows | http://pobox.com/~warmerdam
and watch the world go round - Rush    | Geospatial Programmer for Rent

More information about the gdal-dev mailing list