[BioC] Reading 80 CEL files.

[lgautier at altern.org]

Pure,

The error message tells that at *some point in time* during the execution
of "ReadAffy" a vector of given size cannot be allocated because there is
not enough memory free on your system.

This does not mean that you R session cannot handle anything larger than
309.4 Mb. To check how much memory is currently used by your session, you
can call the garbage collector ("gc()", a side-effect is that the memory
used will be printed.
It is also a good idea to monitor what other processes on your machine are
using significant amounts of memory.

So to get R use more memory, the obvious is to get more RAM (nowadays, 2Gb
is getting quickly limiting) and if get a lot more than 2Gb make sure that
your OS/hardware are capable of making use of it.

If you are only planning to process your data with RMA, the function
"justRMA" is using tricks to use less memory, and should do the job on a
system with 2Gb.

If you really want to look at probe-level data with that much memory,
there are other strategies but they will currently require extra efforts
(I think).


Hoping this helps,


Laurent




>
>>
>> Hello~~~
>>
>>
>>
>> I am trying to read 80 HG-U133A arrays so I did as
>> follows:
>>
>>
>>
>> > memory.limit(size=4095)
>>
>> > options(object.size=10000000000, digits=8,
>> scipen=100, memory=3200483647,
>> contrasts=c("contr.treatment", "contr.poly"))
>>
>> > library(affy)
>>
>> > cel.file <- ReadAffy(celfile.path = ¡¨D://CEL)
>>
>>
>>
>> However, it showed
>>
>>
>>
>> Error: cannot allocate vector of size 309.4 Mb
>>
>>
>>
>> I have tried adding the --max-mem-size=2Gb tag onto
>> my shortcut.
>>
>> The machine has 2G RAM and a 3.30GHz processor.
>>
>> Is there any idea to let R use more memory anywhere?
>>
>> Thank you~
>>
>>
>>
>> Best Regards,
>>
>> Pure
>>
> Pure,
>
> My advice: run them in Linux. We have run up to 1000
> arrays in a linux cluster.
>
> Oscar Puig
>
>
>
>
> _____________________________________________
>> Bioconductor mailing list
>> Bioconductor at stat.math.ethz.ch
>> https://stat.ethz.ch/mailman/listinfo/bioconductor
>> Search the archives:
> http://news.gmane.org/gmane.science.biology.informatics.conductor
>
>
>
>       ____________________________________________________________________________________
> Never miss a thing.  Make Yahoo your home page.
>
> _______________________________________________
> Bioconductor mailing list
> Bioconductor at stat.math.ethz.ch
> https://stat.ethz.ch/mailman/listinfo/bioconductor
> Search the archives:
> http://news.gmane.org/gmane.science.biology.informatics.conductor
>