[R] R and 2 processors

[Liaw, Andy]

On our Linux box (2 P-3 Xeons running 2.4.3 SMP kernel), I can get *some*
operations to use both cpus simultaneously with either threaded ATLAS or MKL
(which is threaded as well).  For example, crossprod(x) can make use of both
cpus.   It's not too clear to me which calls can be threaded and which
can't.  In any case, this is only for linear algebra, not general
computations, so unless sizable chunk of your task can make use of BLAS,
you're not likely to see much gain here.

One this machine, MKL seems to be up to 25% faster than ATLAS.  The strange
thing about threaded ATLAS is that, although it's fast at doing crossprod(),
the speed of other computations are slower than the non-threaded ATLAS.  I
have no idea why this is.

Cheers,
Andy

> -----Original Message-----
> From: Jan de Leeuw [mailto:deleeuw at stat.ucla.edu]
> Sent: Monday, January 21, 2002 9:49 PM
> To: Agustin Lobo
> Cc: r-help
> Subject: Re: [R] R and 2 processors
> 
> 
> On OS X the ATLAS library uses threading (as well as Altivec on G4's).
> Of course the OS is threaded as well.
> 
> On Monday, January 21, 2002, at 02:23 PM, Agustin Lobo wrote:
> 
> >
> > As far as I can see by searching the archieves, R could not
> > take advantage of two processors in the same machine but there
> > were plans for implementing multithreading
> > in R. Is this work going on?
> >
> > On the other hand, if two processors are
> > present in the same linux machine is it possible to focus one
> > processor on tasks of the operative system and the other
> > on R and/or other real computing? Could anyone point me
> > to info on this subject?
> >
> > Agus
> >
> > Dr. Agustin Lobo
> > Instituto de Ciencias de la Tierra (CSIC)
> > Lluis Sole Sabaris s/n
> > 08028 Barcelona SPAIN
> > tel 34 93409 5410
> > fax 34 93411 0012
> > alobo at ija.csic.es
> >
> >
> > 
> -.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.
> -.-.-.-.-.-
> > .-.-.-
> > r-help mailing list -- Read http://www.ci.tuwien.ac.at/~hornik/R/R-
> > FAQ.html
> > Send "info", "help", or "[un]subscribe"
> > (in the "body", not the subject !)  To: 
> r-help-request at stat.math.ethz.ch
> > 
> _._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._.
> _._._._._._
> > ._._._
> >
> >
> ===
> Jan de Leeuw; Professor and Chair, UCLA Department of Statistics;
> US mail: 9432 Boelter Hall, Box 951554, Los Angeles, CA 90095-1554
> phone (310)-825-9550;  fax (310)-206-5658;  email: 
> deleeuw at stat.ucla.edu
> homepage: http://www.stat.ucla.edu/~deleeuw
> ========================================================
>            No matter where you go, there you are. --- Buckaroo Banzai
>                     
> http://www.stat.ucla.edu/~deleeuw/sounds/nomatter.au
> ========================================================
> 
> -.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.
> -.-.-.-.-.-.-.-.-
> r-help mailing list -- Read 
> http://www.ci.tuwien.ac.at/~hornik/R/R-FAQ.html
> Send "info", "help", or "[un]subscribe"
> (in the "body", not the subject !)  To: 
> r-help-request at stat.math.ethz.ch
> _._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._.
> _._._._._._._._._
> 

-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-
r-help mailing list -- Read http://www.ci.tuwien.ac.at/~hornik/R/R-FAQ.html
Send "info", "help", or "[un]subscribe"
(in the "body", not the subject !)  To: r-help-request at stat.math.ethz.ch
_._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._