[R] Error: cannot allocate vector of size 3.4 Gb

Wed Nov 11 15:03:09 CET 2009

Hi Peng,

the major problem about your specific case is that when creating the  
final object, we need to set dimnames() appropriately. This triggers a  
copy of the object and that's where you get the error you describe.

With the current release, unfortunately, there isn't much to do  
(unless you're willing to add more memory). For the next release, I  
have plans of addressing issues like this and reduce memory footprint  
when processing data with the oligo package. It's nothing that can be  
done right now, as it takes some time, but I expect everything to be  
ready for the next release.

If you're trying to run RMA on your data, I can think of ways of  
working around this problem.

Cheers,

b

On Nov 7, 2009, at 5:46 PM, Benilton Carvalho wrote:

> ok, i'll take a look at this and get back to you during the week. b
>
> On Nov 7, 2009, at 1:19 PM, Peng Yu wrote:
>
>> Most of the 8GB was available, when I run the code, because R was the
>> only computation session running.
>>
>> On Sat, Nov 7, 2009 at 7:51 AM, Benilton Carvalho
>> <bcarvalh at jhsph.edu> wrote:
>>> you haven't answered how much resource you have available when you
>>> try
>>> reading in the data.
>>>
>>> with the mouse exon chip, the math is the same i mentioned before.
>>>
>>> having 8 GB, you should be able to read in 70 samples of this chip.
>>> if you
>>> can't, that's because you don't have enough resources when trying
>>> to read.
>>>
>>> best,
>>>
>>> b
>>>
>>> On Nov 7, 2009, at 10:12 AM, Peng Yu wrote:
>>>
>>>> On Fri, Nov 6, 2009 at 8:19 PM, Benilton Carvalho <bcarvalh at jhsph.edu
>>>>>
>>>> wrote:
>>>>>
>>>>> this is converging to bioc.
>>>>>
>>>>> let me know what your sessionInfo() is and what type of CEL files
>>>>> you're
>>>>> trying to read, additionally provide exactly how you reproduce the
>>>>> problem.
>>>>
>>>>
>>>> Here is my sessionInfo(). pname is 'moex10stv1cdf'.
>>>>
>>>>> for (f in list.celfiles('.',full.names=T,recursive=T)) {
>>>>
>>>> +   print(f)
>>>> +   pname=cleancdfname(whatcdf(f))
>>>> +   print(pname)
>>>> + }
>>>>
>>>>
>>>>> sessionInfo()
>>>>
>>>> R version 2.9.2 (2009-08-24)
>>>> x86_64-unknown-linux-gnu
>>>>
>>>> locale:
>>>>
>>>> LC_CTYPE
>>>> =
>>>> en_US
>>>> .UTF
>>>> -8
>>>> ;LC_NUMERIC
>>>> =
>>>> C
>>>> ;LC_TIME
>>>> =
>>>> en_US
>>>> .UTF
>>>> -8
>>>> ;LC_COLLATE
>>>> =
>>>> en_US
>>>> .UTF
>>>> -8
>>>> ;LC_MONETARY
>>>> =
>>>> C
>>>> ;LC_MESSAGES
>>>> =
>>>> en_US
>>>> .UTF
>>>> -8
>>>> ;LC_PAPER
>>>> =
>>>> en_US
>>>> .UTF
>>>> -8
>>>> ;LC_NAME
>>>> =
>>>> C
>>>> ;LC_ADDRESS
>>>> =C;LC_TELEPHONE=C;LC_MEASUREMENT=en_US.UTF-8;LC_IDENTIFICATION=C
>>>>
>>>> attached base packages:
>>>> [1] stats     graphics  grDevices utils     datasets  methods    
>>>> base
>>>>
>>>> other attached packages:
>>>> [1] pd.moex.1.0.st.v1_2.4.1 RSQLite_0.7-2           DBI_0.2-4
>>>> [4] oligo_1.8.3             preprocessCore_1.6.0
>>>> oligoClasses_1.6.0
>>>> [7] Biobase_2.4.1
>>>>
>>>> loaded via a namespace (and not attached):
>>>> [1] affxparser_1.16.0 affyio_1.12.0     Biostrings_2.12.9
>>>> IRanges_1.2.3
>>>> [5] splines_2.9.2
>>>>
>>>>
>>>>> it appears to me, i'm not sure, that you start a fresh session of
>>>>> R and
>>>>> then
>>>>> tries to read in the data - how much resource do you have
>>>>> available when
>>>>> you
>>>>> try reading in the data? having 8GB RAM does not mean that you
>>>>> have 8GB
>>>>> when
>>>>> you tried the task.
>>>>>
>>>>> b
>>>>>
>>>>> On Nov 7, 2009, at 12:08 AM, Peng Yu wrote:
>>>>>
>>>>>> On Fri, Nov 6, 2009 at 5:00 PM, Marc Schwartz <marc_schwartz at me.com
>>>>>>>
>>>>>> wrote:
>>>>>>>
>>>>>>> On Nov 6, 2009, at 4:19 PM, Peng Yu wrote:
>>>>>>>
>>>>>>>> On Fri, Nov 6, 2009 at 3:39 PM, Charlie Sharpsteen
>>>>>>>> <chuck at sharpsteen.net>
>>>>>>>> wrote:
>>>>>>>>>
>>>>>>>>> On Fri, Nov 6, 2009 at 1:30 PM, Peng Yu <pengyu.ut at gmail.com>
>>>>>>>>> wrote:
>>>>>>>>>>
>>>>>>>>>> I run R on a linux machine that has 8GB memory. But R gives
>>>>>>>>>> me an
>>>>>>>>>> error "Error: cannot allocate vector of size 3.4 Gb". I'm
>>>>>>>>>> wondering
>>>>>>>>>> why it can not allocate 3.4 Gb on a 8GB memory machine. How
>>>>>>>>>> to fix
>>>>>>>>>> the
>>>>>>>>>> problem?
>>>>>>>>>
>>>>>>>>> Is it 32-bit R or 64-bit R?
>>>>>>>>>
>>>>>>>>> Are you running any other programs besides R?
>>>>>>>>>
>>>>>>>>> How far into your data processing does the error occur?
>>>>>>>>>
>>>>>>>>> The more statements you execute, the more "fragmented" R's
>>>>>>>>> available
>>>>>>>>> memory pool becomes.  A 3.4 Gb chunk may no longer be
>>>>>>>>> available.
>>>>>>>>
>>>>>>>> I'm pretty sure it is 64-bit R. But I need to double check.  
>>>>>>>> What
>>>>>>>> command I should use to check?
>>>>>>>>
>>>>>>>> It seems that it didn't do anything but just read a lot of  
>>>>>>>> files
>>>>>>>> before it showed up the above errors.
>>>>>>>
>>>>>>>
>>>>>>> Check the output of:
>>>>>>>
>>>>>>> .Machine$sizeof.pointer
>>>>>>>
>>>>>>> If it is 4, R was built as 32 bit, if it is 8, R was built as
>>>>>>> 64 bit.
>>>>>>> See
>>>>>>> ?.Machine for more information.
>>>>>>
>>>>>> It is 8. The code that give the error is listed below. There are
>>>>>> 70
>>>>>> celfiles. I'm wondering how to investigate what cause the
>>>>>> problem and
>>>>>> fix it.
>>>>>>
>>>>>> library(oligo)
>>>>>> cel_files = list.celfiles('.', full.names=T,recursive=T)
>>>>>> data=read.celfiles(cel_files)
>>>>>>
>>>>>>> You can also check:
>>>>>>>
>>>>>>> R.version$arch
>>>>>>>
>>>>>>> and
>>>>>>>
>>>>>>> .Platform$r_arch
>>>>>>>
>>>>>>> which for 64 bit should show x86_64.
>>>>>>>
>>>>>>> HTH,
>>>>>>>
>>>>>>> Marc Schwartz
>>>>>>>
>>>>>>>
>>>>>>
>>>>>> ______________________________________________
>>>>>> 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.
>>>>>
>>>>>
>>>>
>>>> ______________________________________________
>>>> 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.
>>>
>>>
>>
>> ______________________________________________
>> 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.
>
> ______________________________________________
> 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.