[R] re| memory.limit()

ripley@stats.ox.ac.uk ripley at stats.ox.ac.uk
Wed Nov 20 17:24:33 CET 2002

On Wed, 20 Nov 2002, Paul Gilbert wrote:

> ripley at stats.ox.ac.uk wrote:
> >
> > Not quite!
> >
> > memory.limit only exists on Windows, and is not the same as --max-mem-size
> > (which you probably never need to set on Windows).
> >
> > From the CHANGES file for rw1060:
> >
> >   The default memory limit has been raised to the smaller of 1Gb and the
> >   amount of physical RAM.  (See the changes for rw1040 and rw1020 below.)
> Brian
> My limited understand of Windows memory management is that it cannot handle
> programs that are bigger than physical memory. Is my understanding correct, and

No.  It has had virtual memory since ca 1990.

> thus the reason for the above limit? For this and other reasons I have tended to
> stick with Unix/Linux versions of R, and I am wondering if this should be seen
> as a Windows limitation or a limitation imposed by R in Windows.

Windows is very slow at using more that the physical RAM, especially on
16-bit Windows (which 95/98/ME are for this purpose).  But it can do it.

The reason for the 1Gb limit is to do with the limited total space Windows
gives (2Gb or 3Gb across all processes).

Yes, Linux on the same machine handles large amounts of memory better.
But the gap is much less than it used to be.


Brian D. Ripley,                  ripley at 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

r-help mailing list -- Read http://www.ci.tuwien.ac.at/~hornik/R/R-FAQ.html
Send "info", "help", or "[un]subscribe"
(in the "body", not the subject !)  To: r-help-request at stat.math.ethz.ch

More information about the R-help mailing list