S-PLUS on multi-processor machines [was: S+ and Bi processor machine]

Prof Brian Ripley (ripley@stats.ox.ac.uk)
Sun, 1 Feb 1998 09:16:46 +0000 (GMT)


Waitung Ho wrote:
>
> To make use of more than one cpu, you need to have a multi-thread OS
> (e.g. win NT) and a multi-thread software. However, splus doen't seem to
> be a multi-thread software because there is only one thread assocated
> with splus.
>
> From Waitung
>
> On Sat, 31 Jan 1998, Badih.Ghattas wrote:
>
> > Hello every body ,
> >
> > does any body have any experience using S+ on a Bi processor machine ??
> > Does this depend only on the system used ?
> > More precisely, is there a real gain if i add to my PC (266MHZ) a second
> > processor ??
> >
> >
> > THANKS.
> >

Similar considerations apply to Unix, but in both cases it is a bit more
complicated than multiple threads. Windows NT and Unix (e.g. Solaris) are
multi-tasking operating systems and can (and do) run multiple tasks
simultaneously on multiple processors. So one immediate benefit of two
processors is that system tasks do not need to interrupt the S-PLUS
application; another is that the machine may be usable for other tasks
(editing, for example).

Using more than one processor is a form of parallelization, and this needs
thought for most statistical computations. I often use a bulk form of
parallelization, for example by running two or more folds of a K-fold
cross-validation experiment at the same time. If you do run more than
one copy of S-PLUS it is easiest to ensure that they have separate working
directories. I would have thought that almost all large statistical
computations could be re-organized into independent pieces to be run on
separate copies of S-PLUS.

My understanding is that Windows 95 is both multi-threaded and
multi-tasking: it just does not have support for more than one processor.
There were (at least) Unices like that: you need a multi-processor OS.
They differ markedly in their effectiveness.

-- 
Brian D. Ripley,                  ripley@stats.ox.ac.uk
Professor of Applied Statistics,  http://www.stats.ox.ac.uk/~ripley/
University of Oxford,             Tel:  +44 1865 272861 (self)
1 South Parks Road,                     +44 1865 272860 (secr)
Oxford OX1 3TG, UK                Fax:  +44 1865 272595