[R] R on fairly large machines

[Peter Dalgaard BSA]

Allan Tingey <al at genetics.utah.edu> writes:

> Hi,
> 
> We are considering using R on a large problem that will require four
> 64 bit cpu's and 64Gb of ram and some flavor of unix.  We are just
> wondering if R is going to be happy in this environment.  If have
> experience with this please let me know.

You're probably going into uncharted territory at least to some
extent, but various people have been gathering experiences with 64-bit
platforms. 

I'm not rich enough to have hands-on experience, but my impression is
that things are quite dependent on the precise platform. The free
software OS's are generally quite well standardized so that you can
use (near-)standard procedures to get going. The commercial ones will
require some elbow grease to get compiler switches and library
settings right, and to get various support libraries (readline, jpeg,
bz, etc.) in place. 

Then again, there are preciously few platforms that support that kind
of memory size - I know that Sun will, but the speeds that I have seen
from their CPUs have not been impressive.

-- 
   O__  ---- Peter Dalgaard             Blegdamsvej 3  
  c/ /'_ --- Dept. of Biostatistics     2200 Cph. N   
 (*) \(*) -- University of Copenhagen   Denmark      Ph: (+45) 35327918
~~~~~~~~~~ - (p.dalgaard at biostat.ku.dk)             FAX: (+45) 35327907