[GRASS-user] 64 bit or parallel processing

Charles Ehlschlaeger c.ehlschlaeger at insightbb.com
Tue Feb 27 08:30:21 EST 2007


The inherent nature of raster data makes it very easy to implement most
raster commands in a parallel processing environment. Back in the early
1990's, researchers at Oak Ridge National Laboratory wanted to use some of
my research in spatial data uncertainty modeling. In less than two hours, an
ORNL programmer and I identified the functions that could be multi-threaded.
The programmer was experienced with the parallel processing C modules ORNL
used in its super computer. By the next day, the programmer had my programs
working on the super computer.

A quick google search found OdinMP, http://fenixforge.com/projects/odinmp,
as a possible free open source parallel processer using the OpenMP
future-standard.

Chuck Ehlschlaeger, Associate Professor & GIS Center Director
Department  of  Geography,      Western  Illinois  University
215 Tillman Hall,   1 University Circle,  Macomb,  IL   61455
cre111 at wiu.edu,    phone: 309-298-1841,     fax: 309-298-3003

-----Original Message-----
From: grassuser-bounces at grass.itc.it [mailto:grassuser-bounces at grass.itc.it]
On Behalf Of francesco.pirotti
Sent: Tuesday, February 27, 2007 1:38 AM
To: grassuser at grass.itc.it
Subject: [GRASS-user] 64 bit or parallel processing

Dear Grass users,

Has anyone experimented using 64 bit machines or some kind of parallel 
processing for some GRASS modules?  Is it a possible thing?  Does GRAS have 
any modules prepared for such conditions?  I am not a superexpert, I was 
wondering if anyone can give me clues on how can operations be made faster 
using special hardware... for example is I can significantly speed up 
spline interpolation using parallel machines, or clusters (I am shooting in 
the dark, since I am not even sure what clusters are )

Thanks guys,

Francesco Pirotti

_______________________________________________
grassuser mailing list
grassuser at grass.itc.it
http://grass.itc.it/mailman/listinfo/grassuser




More information about the grass-user mailing list