R help - Sep 2012 - memory use problem.

I having issues with my memory size in R for Windows.  I have implemented the
3gb switch for Windows 7 32 bit with 8gb of memory, only 4gb is available
for use due to 32 bit limitations.  512Mb is taken by the shared video card,
the memory limit is 2559mb yet I cannot use more than 1.5 gb of memory.  I
have searched eveywhere and cannot find any resolution to my issue.

> memory.limit()
[1] 2559
> memory.size()
[1] 399.47
> a = aov(data$cv~(data$Lo+data$La+data$Mo+data$Co))
Error: cannot allocate vector of size 1.4 Gb
> gc()
           used  (Mb) gc trigger  (Mb) max used  (Mb)
Ncells   726003  19.4    2251281  60.2  1677683  44.8
Vcells 37249684 284.2   93398451 712.6 93388743 712.5
> (1024 * 14) + 44.8 + 712.5
[1] 15093.3
> 
Thank you,

Chris



--
View this message in context:
http://r.789695.n4.nabble.com/memory-use-problem-tp4642658.html
Sent from the R help mailing list archive at Nabble.com.

On 10/09/2012 16:13, chrisk wrote:
> I having issues with my memory size in R for Windows.  I have implemented
the
> 3gb switch for Windows 7 32 bit with 8gb of memory, only 4gb is available
> for use due to 32 bit limitations.  512Mb is taken by the shared video
card,
> the memory limit is 2559mb yet I cannot use more than 1.5 gb of memory.  I
> have searched eveywhere and cannot find any resolution to my issue.
?'Memory-limits' told you why

      It is not normally possible to allocate as much as 2Gb to a single
      vector in a 32-bit build of R even on 64-bit Windows because of
      preallocations by Windows in the middle of the address space.

'Search eveywhere' seems not to have included the R help ... I agree 
that Microsoft is very coy about what it does to your address space.
>
>
>> memory.limit()
> [1] 2559
>> memory.size()
> [1] 399.47
>> a = aov(data$cv~(data$Lo+data$La+data$Mo+data$Co))
> Error: cannot allocate vector of size 1.4 Gb
>> gc()
>             used  (Mb) gc trigger  (Mb) max used  (Mb)
> Ncells   726003  19.4    2251281  60.2  1677683  44.8
> Vcells 37249684 284.2   93398451 712.6 93388743 712.5
>> (1024 * 14) + 44.8 + 712.5
> [1] 15093.3
>>
>
> Thank you,
>
> Chris
>

-- 
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