[Rd] Allocated Memory Warnings in Vista 32 bit with 4 GB (PR#12557)

[drmark at marktab.com]

Hello

I am running Windows Vista 32 with 4 GB (installed, though Windows of cours=
e only recognizes 3326 MB, as reported by Windows "My Computer")

I am running R 2.7.1

I was trying to read in a comma delimited single column CSV file, assign th=
at file to a variable ("data") and then extract a sample (assigned to "part=
ial").  I was getting memory allocation errors, and from the log below, I s=
tarted to see a pattern which indicates that the "error" is related to how =
R views memory allocation, and perhaps how R is using or reading memory on =
my specific system.  The error, I would guess, is probably related to how R=
 is reporting the memory available.

My source files have the same number of rows (or cases) as the name of the =
file.  Thus, data10000 has 10,000 observations, and data300000 has 300,000 =
observations.

Here is my history which shows the problem, and inconsistency in memory all=
ocation reporting:

  > data=3Dread.csv("data10000.csv",header=3DFALSE)
  > partial=3Dsample(data,5000,T)
  > data=3Dread.csv("data100000.csv",header=3DFALSE)
  Error in file(file, "r") : cannot open the connection
  In addition: Warning message:
  In file(file, "r") :
    cannot open file 'data100000.csv': No such file or directory
  > data=3Dread.csv("data300000.csv",header=3DFALSE)
  > partial=3Dsample(data,5000,T)
  Error: cannot allocate vector of size 2.3 Mb
  In addition: Warning messages:
  1: In names(y) <- make.unique(cols) :
    Reached total allocation of 1535Mb: see help(memory.size)
  2: In names(y) <- make.unique(cols) :
    Reached total allocation of 1535Mb: see help(memory.size)
  3: In names(y) <- make.unique(cols) :
    Reached total allocation of 1535Mb: see help(memory.size)
  4: In names(y) <- make.unique(cols) :
    Reached total allocation of 1535Mb: see help(memory.size)
  > partial=3Dsample(data,5,T)
  Error: cannot allocate vector of size 2.3 Mb
  In addition: Warning messages:
  1: In names(y) <- make.unique(cols) :
    Reached total allocation of 1535Mb: see help(memory.size)
  2: In names(y) <- make.unique(cols) :
    Reached total allocation of 1535Mb: see help(memory.size)
  3: In names(y) <- make.unique(cols) :
    Reached total allocation of 1535Mb: see help(memory.size)
  4: In names(y) <- make.unique(cols) :
    Reached total allocation of 1535Mb: see help(memory.size)
  > partial=3Dsample(data,1,T)
  Warning messages:
  1: In attributes(.Data) <- c(attributes(.Data), attrib) :
    Reached total allocation of 1535Mb: see help(memory.size)
  2: In attributes(.Data) <- c(attributes(.Data), attrib) :
    Reached total allocation of 1535Mb: see help(memory.size)
  > partial=3Dsample(data,1,T)
  > partial=3Dsample(data,5000,T)
  Error: cannot allocate vector of size 2.3 Mb
  In addition: Warning messages:
  1: In names(y) <- make.unique(cols) :
    Reached total allocation of 1535Mb: see help(memory.size)
  2: In names(y) <- make.unique(cols) :
    Reached total allocation of 1535Mb: see help(memory.size)
  3: In names(y) <- make.unique(cols) :
    Reached total allocation of 1535Mb: see help(memory.size)
  4: In names(y) <- make.unique(cols) :
    Reached total allocation of 1535Mb: see help(memory.size)
  > partial[1:5,]
  [1]  0.3204279  1.6583593 -0.3456585  1.2951363 -1.1096974
  > rm(partial)
  > partial=3Dsample(data,5000,T)
  Error: cannot allocate vector of size 2.3 Mb
  In addition: Warning messages:
  1: In names(y) <- make.unique(cols) :
    Reached total allocation of 1535Mb: see help(memory.size)
  2: In names(y) <- make.unique(cols) :
    Reached total allocation of 1535Mb: see help(memory.size)
  3: In names(y) <- make.unique(cols) :
    Reached total allocation of 1535Mb: see help(memory.size)
  4: In names(y) <- make.unique(cols) :
    Reached total allocation of 1535Mb: see help(memory.size)
  5: In names(y) <- make.unique(cols) :
    Reached total allocation of 1535Mb: see help(memory.size)
  6: In names(y) <- make.unique(cols) :
    Reached total allocation of 1535Mb: see help(memory.size)
  > partial[1:5,]
  Error: object "partial" not found
  > partial=3Dsample(data,1,T)
  Error: cannot allocate vector of size 2.3 Mb
  In addition: Warning messages:
  1: In attributes(.Data) <- c(attributes(.Data), attrib) :
    Reached total allocation of 1535Mb: see help(memory.size)
  2: In attributes(.Data) <- c(attributes(.Data), attrib) :
    Reached total allocation of 1535Mb: see help(memory.size)
  3: In attributes(.Data) <- c(attributes(.Data), attrib) :
    Reached total allocation of 1535Mb: see help(memory.size)
  4: In attributes(.Data) <- c(attributes(.Data), attrib) :
    Reached total allocation of 1535Mb: see help(memory.size)
  > partial=3Dsample(data,1,T)
  Warning messages:
  1: In attributes(.Data) <- c(attributes(.Data), attrib) :
    Reached total allocation of 1535Mb: see help(memory.size)
  2: In attributes(.Data) <- c(attributes(.Data), attrib) :
    Reached total allocation of 1535Mb: see help(memory.size)
  > partial=3Dsample(data,1,T)
  > partial=3Dsample(data,5000,T)
  Error: cannot allocate vector of size 2.3 Mb
  In addition: Warning messages:
  1: In names(y) <- make.unique(cols) :
    Reached total allocation of 1535Mb: see help(memory.size)
  2: In names(y) <- make.unique(cols) :
    Reached total allocation of 1535Mb: see help(memory.size)
  3: In names(y) <- make.unique(cols) :
    Reached total allocation of 1535Mb: see help(memory.size)
  4: In names(y) <- make.unique(cols) :
    Reached total allocation of 1535Mb: see help(memory.size)
  >=20


I have been able to reproduce the situation several times.  For example, wh=
at is curious is that a command like partial=3Dsample(data,1,T) may or may =
not produce a warning message.  In the history, obviously 2.3MB is well bel=
ow the allocated 1535MB.  I had been getting this error on my original larg=
er source data (1,000,000 observations), but since I am a new R user, it wa=
s not clear to me what these errors mean.

If you need the original source files, or have additional questions, please=
 let me know.  I am a certified .NET software developer, and although I do =
not know everything about Windows development, I believe my abilities are a=
bove average.



Mark Tabladillo=20
Atlanta, GA
marktab at email.phoenix.edu
Alternate:  x2008 at marktab.com


	[[alternative HTML version deleted]]