If you need more than 2^31 - 1, have a look at bigmemory as a possibility
(particularly if you aren't shy about C++ and want to work directly with
the object).

Jay

On Thu, Aug 9, 2012 at 9:23 AM, Mathias Bader <[email protected]> wrote:

>
> Hello Davor
>
> Thank you for your message.
>
> I came up with similar numbers: 4.29 billion elements for unsigned int. My
> simulation goes easily to one million in some seconds, so I might reach
> that number, but I think that is not the reason why my program crashes.
>
> I think the problem has something to do with the List-creation of Rcpp,
> someone had a similar error:
>     http://lists.r-forge.r-**project.org/pipermail/rcpp-**
> devel/2012-March/003657.html<http://lists.r-forge.r-project.org/pipermail/rcpp-devel/2012-March/003657.html>
> Does anyone know whether there is a workaround for this problem?
>
> Mathias
>
>
>
>
>
> On Thu, 09 Aug 2012 06:16:03 -0700, Davor Cubranic wrote:
>
>> On 12-08-09 05:08 AM, Mathias Bader wrote:
>>
>>> My question: Is there any restriction on the size of the vectors which I
>>> hand from C++ to R? Because during the MCMC simulation the vectors might
>>> become really big.
>>>
>>
>> How big are we talking about? R uses 32-bit ints for indexes, which
>> would be over 2 billion elements.
>>
>> What is the crash message? Segmentation fault? Have you tried running
>> with valgrind to check for memory allocation problems?
>>
>> Davor
>>
>
> ______________________________**_________________
> Rcpp-devel mailing list
> [email protected]**project.org<[email protected]>
> https://lists.r-forge.r-**project.org/cgi-bin/mailman/**
> listinfo/rcpp-devel<https://lists.r-forge.r-project.org/cgi-bin/mailman/listinfo/rcpp-devel>
>



-- 
John W. Emerson (Jay)
Associate Professor of Statistics, Adjunct, and Acting Director of Graduate
Studies
Department of Statistics
Yale University
http://www.stat.yale.edu/~jay
_______________________________________________
Rcpp-devel mailing list
[email protected]
https://lists.r-forge.r-project.org/cgi-bin/mailman/listinfo/rcpp-devel

Reply via email to