[R-SIG-Mac] How to Speed up R on the G5

Byron Ellis ellis at stat.harvard.edu
Tue Feb 8 10:32:40 CET 2005

Interesting. I was under the impression that it was parallel in the 
SIMD sense, not in the processor sharing sense. We just got mini XServe 
cluster (sadly, my proposal to call it McNugget was rejected) so I'll 
have to check it out. Its still running the stock iNquiry R (1.9.1 
IIRC), but thats easy enough to change ;-)

On Feb 8, 2005, at 12:38 AM, Douglas Bates wrote:

> Regarding the question of whether veclib is parallelized, I'm 
> reasonably sure that it is.  A few months ago I compiled a version of 
> R using veclib for the BLAS and ran a simple timing test on a matrix 
> crossproduct.
> mm <- matrix(rnorm(9e6), nc = 3000)
> system.time(crossprod(mm), gc = TRUE)
> system.time(crossprod(mm), gc = TRUE)
> system.time(crossprod(mm), gc = TRUE)
> I found on a dual processor system that the elapsed time was about 
> half the user CPU time.
> I don't have access to that machine right now but I should be able to 
> do such a test again next week or perhaps Michael Redmond can do it 
> this week.
> _______________________________________________
> R-SIG-Mac mailing list
> R-SIG-Mac at stat.math.ethz.ch
> https://stat.ethz.ch/mailman/listinfo/r-sig-mac
Byron Ellis (ellis at stat.harvard.edu)
"Oook" -- The Librarian

More information about the R-SIG-Mac mailing list