[R] best quad core configuration?

Vincent Goulet vincent.goulet at act.ulaval.ca
Wed Jul 30 19:49:05 CEST 2008


Le mer. 30 juil. à 13:04, Alan Swanson a écrit :

> I am purchasing a quad-core 64-bit system.  My proposed  
> configuration is an
> intel core duo quad with 8Gb memory and an ubuntu OS.  I'm choosing  
> ubunto
> because I have good local support, but it doesn't seem as well  
> supported as
> debian for R.

And what makes you think that? First off, Ubuntu is Debian based, so  
every Debian package (well, the R ones at least) is also in Ubuntu.  
Second, we also provide backports of R for Ubuntu releases down to  
Dapper. I consider this is rather good support; see

	http://cran.r-project.org/bin/linux/ubuntu/

>  Will ubuntu limit me in any significant way?   I have noticed
> a bias towards AMD machines among R users - are there compelling  
> reasons to
> go with an equivalent AMD system instead?  Will I be able to use amd64
> binaries on this system, or would I have to compile everything myself?

As the README file linked above mentions, we provide up-to-date R  
binaries for i386 and amd64.

For your other points, I will let more competent people answer.

---
   Vincent Goulet, Associate Professor
   École d'actuariat
   Université Laval, Québec
   Vincent.Goulet at act.ulaval.ca   http://vgoulet.act.ulaval.ca

>
> Would it difficult to install Intel MKL 10.0 BLAS and LAPACK  
> libraries on
> this machine?  Would it be difficult to run both 32-bit and 64-bit  
> versions
> of R on this machine?  The most computationally-intensive thing I  
> will be
> using the machine for is large MCMC simulations, probably using  
> multiple
> instances of LinBUGS.  Any advice on configuring by machine for  
> maximum
> speed for these simulations?
>
> Thanks,
>
> Alan Swanson
>
>
>
>
> 	[[alternative HTML version deleted]]
>
> ______________________________________________
> R-help at r-project.org mailing list
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
> and provide commented, minimal, self-contained, reproducible code.



More information about the R-help mailing list