[GRASS-dev] Multi-threading and use of multiple processors

[Glynn Clements]

Colin Nielsen wrote:

> A colleague of mine would like to run a cost analysis (with either
> r.walk or r.cost) on a very very large raster, ~ 1500 km x 800 km with
> about 30 m resolution equals about 40 billion cells. This is obviously
> going to take a ridiculously long time, and he would like to make use
> of his dual quad-core processors to speed up this process (he has
> decided that a week of running time is acceptable).
> 
> Is there any plan to add multi-threading to these or other modules (I
> think I remember something about r.mapcalc and another a while back)?
> What would it take to do this? Is there anything else anyone can
> recommend to help speed up this analysis (I think I remember something
> about keeping the whole thing in memory as well...)?

Unfortunately, adding parallelism to these modules wouldn't be
straightfoward, as you can't easily divide the data up into isolated
portions.

The simplest way to speed them up would be to replace the use of the
segment library with a more efficient version, e.g. that used in
r.proj.seg (r.proj in 7.0).

-- 
Glynn Clements <glynn at gclements.plus.com>