[R] The snow package

[A.J. Rossini]

njmulakk <njmulakk at uclink.berkeley.edu> writes:

> I have a few questions regarding the snow package:
>
> 1) Is fine-grained parallelism even possible with R/snow?
>
> (ie, one task that spawns sub-tasks to do work, with results sent back to the 
> parent task).
>
> The example(s) only seem to be totally-independent processing (which can be 
> done just as easily
> with shell-script wrappers....)

You can use the underlying tools (rpvm or Rmpi) to do it.  I think
sockets, as well, but that might take slightly more thought.

> 2) If so, is an example available?

Not readily packaged up.

> 3. If not, then it is still worth messing with snow to see if you can get an R 
> program to skip through multiple directories, doing the independent processing 
> in each to avoid a potential race condition that could occur if multiple R 
> process one with one .RData file.

Sure.  Or just make sure to load/save to other places besides .RData.

?save will help on that.

best,
-tony

-- 
A.J. Rossini     			
rossini at u.washington.edu            http://www.analytics.washington.edu/ 
Biomedical and Health Informatics   University of Washington
Biostatistics, SCHARP/HVTN          Fred Hutchinson Cancer Research Center
UW   :              FAX=206-543-3461 | moving soon to a permanent office
FHCRC: 206-667-7025 FAX=206-667-4812 | Voicemail is pretty sketchy/use Email

CONFIDENTIALITY NOTICE: This e-mail message and any attachme...{{dropped}}