[BioC] bioconductor on xserve

[Colin Smith]

Guoneng-

I haven't run Bioconductor scripts on Xserve hardware specifically. I 
have however successfully set up a test installation of webbioc on a 
small cluster of a G3 and G4 Power Mac linked together with the Sun 
Grid Engine.

Your best bet is probably installing the prebuilt R package (from CRAN) 
onto each cluster node. Then you'll need to create a R library 
directory where you install all the Bioconductor packages. That 
directory could either be shared over NFS, etc. or copied to each node. 
It's fairly straightforward to write simple shell scripts which can 
execute a file containing a series of R commands. For analysis of 
Affymetrix data, webbioc takes care of all of that for you and has an 
easy to use web interface for submitting jobs to the cluster.

If you only have a user account on the Xserve cluster, it is of course 
possible to compile and run R/Bioconductor from entirely within your 
home directory.

-Colin

On Dec 4, 2004, at 3:00, Guoneng Zhong <guoneng.zhong at yale.edu> wrote:

> Date: Fri, 03 Dec 2004 14:43:23 -0500
> From: Guoneng Zhong <guoneng.zhong at yale.edu>
> Subject: [BioC] bioconductor on xserve
> To: bioconductor at stat.math.ethz.ch
>
> Hi,
>
> Not sure if this is more appropriate for R general or Xserve, but has 
> anyone
> managed to run bioconductor scripts on a cluster of Xserves?
>
> thanks,
> g