R help - Mar 2008 - Cannot allocate large vectors (running out of memory?)

Hi.
   As shown in the simplified example below, I'm having trouble allocating
   memory for large vectors, even though it would appear that there is more
   than enough memory available. That is, even with a memory limit of 1500 MB,
   R 2.6.1 (Win) will allocate memory for a first vector of 285 MB, but not for
   a second vector of the same size. Forcing garbage collection does not seem
   to solve the problem.
   Can anyone explain why is this happening, and how to fix it?
   Thanks,
   Ronnen.
   P.S. E-mailed CCs of posted replies would be appreciated.
   > rm(list=ls(all=TRUE))
   > gc()
            used (Mb) gc trigger  (Mb)  max used  (Mb)
   Ncells 143465  3.9     350000   9.4    350000   9.4
   Vcells  88573  0.7   50380943 384.4 131023877 999.7
   > memory.limit()
   [1] 1535.875
   > n <- 8640 * 4320
   > x=rep(1/3, n)
   > memory.size()
   [1] 578.8543
   > gc()
              used  (Mb) gc trigger  (Mb)  max used  (Mb)
   Ncells   143471   3.9     350000   9.4    350000   9.4
   Vcells 37413375 285.5   78720219 600.6 131023877 999.7
   > y=rep(1/7, n)
   Error: cannot allocate vector of size 284.8 Mb
   > memory.size()
   [1] 578.8543
   > gc()
              used  (Mb) gc trigger  (Mb)  max used  (Mb)
   Ncells   143471   3.9     350000   9.4    350000   9.4
   Vcells 37413375 285.5   78720219 600.6 131023877 999.7
   > version
                  _
   platform       i386-pc-mingw32
   arch           i386
   os             mingw32
   system         i386, mingw32
   status
   major          2
   minor          6.1
   year           2007
   month          11
   day            26
   svn rev        43537
   language       R
   version.string R version 2.6.1 (2007-11-26)

   --
   Ronnen Levinson, Ph.D.
   scientist, Lawrence Berkeley National Lab
   The  Onion  horoscope:  Pisces February 19 - March 20 You will soon be
   unwillingly forced into a flurry of activity when you are chosen to host the
   2014 Winter Olympiad.

See ?"Memory-limits" and rw-FAQ Q2.9.  Since you have already used
more
than 1Gb of what is presumably a 2Gb address space (we don't know your 
Windows version), it is not surprising that there is no 285Mb hole left.

On Mon, 24 Mar 2008, Ronnen Levinson wrote:
>
>   Hi.
>   As shown in the simplified example below, I'm having trouble
allocating
>   memory for large vectors, even though it would appear that there is more
>   than enough memory available. That is, even with a memory limit of 1500
MB,
>   R 2.6.1 (Win) will allocate memory for a first vector of 285 MB, but not
for
>   a second vector of the same size. Forcing garbage collection does not
seem
>   to solve the problem.
>   Can anyone explain why is this happening, and how to fix it?
>   Thanks,
>   Ronnen.
>   P.S. E-mailed CCs of posted replies would be appreciated.
>   > rm(list=ls(all=TRUE))
>   > gc()
>            used (Mb) gc trigger  (Mb)  max used  (Mb)
>   Ncells 143465  3.9     350000   9.4    350000   9.4
>   Vcells  88573  0.7   50380943 384.4 131023877 999.7
>   > memory.limit()
>   [1] 1535.875
>   > n <- 8640 * 4320
>   > x=rep(1/3, n)
>   > memory.size()
>   [1] 578.8543
>   > gc()
>              used  (Mb) gc trigger  (Mb)  max used  (Mb)
>   Ncells   143471   3.9     350000   9.4    350000   9.4
>   Vcells 37413375 285.5   78720219 600.6 131023877 999.7
>   > y=rep(1/7, n)
>   Error: cannot allocate vector of size 284.8 Mb
>   > memory.size()
>   [1] 578.8543
>   > gc()
>              used  (Mb) gc trigger  (Mb)  max used  (Mb)
>   Ncells   143471   3.9     350000   9.4    350000   9.4
>   Vcells 37413375 285.5   78720219 600.6 131023877 999.7
>   > version
>                  _
>   platform       i386-pc-mingw32
>   arch           i386
>   os             mingw32
>   system         i386, mingw32
>   status
>   major          2
>   minor          6.1
>   year           2007
>   month          11
>   day            26
>   svn rev        43537
>   language       R
>   version.string R version 2.6.1 (2007-11-26)
>
>   --
>   Ronnen Levinson, Ph.D.
>   scientist, Lawrence Berkeley National Lab
>   The  Onion  horoscope:  Pisces February 19 - March 20 You will soon be
>   unwillingly forced into a flurry of activity when you are chosen to host
the
>   2014 Winter Olympiad.
> ______________________________________________
> R-help at r-project.org mailing list
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide
http://www.R-project.org/posting-guide.html
> and provide commented, minimal, self-contained, reproducible code.
>
-- 
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 272866 (PA)
Oxford OX1 3TG, UK                Fax:  +44 1865 272595