Step 1 - write some spatial benchmarks. I would benchmark: spatial object creation from files (shapefiles, typically, but this probably more an I/O benchmark), spatial object creation from primitive objects (the SpatialPolygons(list(Polygons(list(matrix))))) stuff), over, and everything in rgeos, as many kriging packages as you can find, and as much of raster as you can think of. Quite a bit of that is already in C so might not get any faster in pqR. But a spatial benchmark suite would be nice anyway. I've just benchmarked my time and I have 0s to do this. On Thu, Jun 27, 2013 at 4:09 PM, Boulanger, Yan
<Yan.Boulanger at rncan-nrcan.gc.ca> wrote:
Dear forumities, Did anyone tried pqR for spatial analyses ? I don't know if the worth the case (I will have to try myself) but it seems that pqR (which was released last week) is able to perform operations in parallel, especially on matrices so I guess it could substantially reduce the computation time when e.g., doing simple mathematical operations on rasters. "One notable improvement is that pqR is able to do some numeric computations in parallel with each other, and with other operations of the interpreter, on systems with multiple processors or processor cores" (http://radfordneal.wordpress.com/2013/06/22/announcing-pqr-a-faster-version-of-r/) The program is only available for the moment, for UNIX and Linux platforms although it seems to run correctly on Mac. Any thoughts? Yan Yan Boulanger, Postdoctoral Visiting Fellow Ressources Naturelles Canada, Canadian Forest Service Centre de Foresterie des Laurentides 1055, rue du P.E.P.S. C.P. 10380, succ. Sainte-Foy Qu?bec (Qu?bec) Canada G1V 4C7 Tel. : +1 418 649-6859 [[alternative HTML version deleted]]