[ESS] Connect xemacs/ess to an existing R process to use Rmpi?

[Ramon Diaz-Uriarte]

Dear Adam,

This might not be of much use, but just in case. Under GNU/Linux, and
with LAM/MPI everything works as usual. I open a shell from emacs,
start the LAM/MPI universe (lamboot), and then M-X R. And then work as
usual (library(Rmpi); mpi.spawn.Rslaves(...); etc, etc).

Best,

R.

On Dec 18, 2007 11:43 PM, Adam Wilson <adam.wilson at uconn.edu> wrote:
> Greetings all,
>
> I am working with the Rmpi package (
> http://cran.r-project.org/src/contrib/Descriptions/Rmpi.html) to take
> advantage of my multi-core processor to run parallel MCMC chains.  To use
> the rmpi package you need to start the R process from the MPICH2 (
> http://www.mcs.anl.gov/research/projects/mpich2/) utility.  If you just open
> Rgui or Rterm.exe, Rmpi will not function.  See
> http://www.stats.uwo.ca/faculty/yu/Rmpi/ for more details.
>
> Typically, starting MPICH2 and R is done with a shortcut to:
> "C:\Program Files (x86)\MPICH2\bin\mpiexec.exe" -n 1 -localhost -noprompt
> "C:\Program Files (x86)\R\R-2.6.1\bin\Rgui.exe"
>
> My question: is there a way to access a MPI-enabled Rterm process from
> xemacs/ess?  I imagine two possibilities:
> 1) connect xemacs/ess to an existing R process that I've started with
> MPICH2, or
> 2) direct xemacs/ess to start MPICH2 (using some form of the above shortcut)
> and connect to the resulting R process.
>
> I've not been successful in finding any information on either of these
> options in the archives or user manual.
>
> Details on my system are below.
>
> Thanks for any help,
>
> Adam
>
>
> ###################################
>
> I am running R 2.6.1 in 64-bit Windows XP SP2 on a quad-core intel xeon
> 3.0GHz processor.
> XEmacs v21.4.21
> Ess v5.3.6
>
> > Sys.info()
>                       sysname                       release
>                     "Windows"                      "NT 5.2"
>                       version
> "(build 3790) Service Pack 2"
>                       machine
>                         "x86"
>
>
> --
> Adam Wilson
> http://hydrodictyon.eeb.uconn.edu/people/wilson/
> Department of Ecology and Evolutionary Biology
> BioPharm 223
> University of Connecticut
> Tel: 860.486.4157
> Adam.Wilson at UConn.edu
>
>         [[alternative HTML version deleted]]
>
> ______________________________________________
> ESS-help at stat.math.ethz.ch mailing list
> https://stat.ethz.ch/mailman/listinfo/ess-help
>



-- 
Ramon Diaz-Uriarte
Statistical Computing Team
Structural Biology and Biocomputing Programme
Spanish National Cancer Centre (CNIO)
http://ligarto.org/rdiaz