[R] Why was my R process killed spontaneously?

Jeff Newmiller jdnewmil at dcn.davis.CA.us
Fri Aug 24 15:30:50 CEST 2012


Yes, the resources operating systems usually kill for are memory.

No, you have not provided a reproducible example. In general, you need to do something different, such as choose a different algorithm, run on a different computer, or split your problem into smaller pieces.
---------------------------------------------------------------------------
Jeff Newmiller                        The     .....       .....  Go Live...
DCN:<jdnewmil at dcn.davis.ca.us>        Basics: ##.#.       ##.#.  Live Go...
                                      Live:   OO#.. Dead: OO#..  Playing
Research Engineer (Solar/Batteries            O.O#.       #.O#.  with
/Software/Embedded Controllers)               .OO#.       .OO#.  rocks...1k
--------------------------------------------------------------------------- 
Sent from my phone. Please excuse my brevity.

flora flora <floraflora196 at gmail.com> wrote:

>Thanks for your reply.
>
>Does what you said by resources mean memory or something else? Inside
>my
>loop, I removed the objects that are created but not used by the next
>loop
>and did garbage collection as well. Do you have any idea how I should
>modify my code such that the system won't kill it? Thanks.
>
>
>
>
>
>On Thu, Aug 23, 2012 at 5:06 PM, Jeff Newmiller
><jdnewmil at dcn.davis.ca.us>wrote:
>
>> Yes. Operating systems kill processes that consume excessive
>resources.
>> That excess may arise from one large computation or from a small one
>on top
>> of many other allocations... the "straw that broke the camel's back"
>> problem.
>>
>---------------------------------------------------------------------------
>> Jeff Newmiller                        The     .....       .....  Go
>Live...
>> DCN:<jdnewmil at dcn.davis.ca.us>        Basics: ##.#.       ##.#.  Live
>> Go...
>>                                       Live:   OO#.. Dead: OO#.. 
>Playing
>> Research Engineer (Solar/Batteries            O.O#.       #.O#.  with
>> /Software/Embedded Controllers)               .OO#.       .OO#. 
>rocks...1k
>>
>---------------------------------------------------------------------------
>> Sent from my phone. Please excuse my brevity.
>>
>> flora flora <floraflora196 at gmail.com> wrote:
>>
>> >I tried to use gpuCor function in the gputools package of R to
>calcuate
>> >the
>> >pairwise correlations of a matrix of 40,000 columns.
>> >
>> >Becuase there would be memory issues if I use the whole matrix at a
>> >time, I
>> >splitted the matrix into submatrix of 10,000 columns and then
>calculate
>> >the
>> >pairwise correlation of different submatrices. There are altogether
>4
>> >submatrices, so I need to calculate the pearson correlation of sub
>> >matrix 1
>> >with sub matrix 1,2,3,4, and sub matrix 2 with submatrix 1,2,3,4,
>etc.
>> >
>> >The program runs well at first, but at the last step, which is
>> >calculating
>> >the correlation between submatrix 4 with itself, the program was
>killed
>> >and
>> >gave no error messages.
>> >
>> >Have anybody else encountered this before?
>> >
>> >Actually it doesn't have to be related with gpuCor. Just generally
>> >speaking, in what circumstances would a R program be killed
>> >spontaneously
>> >without any error messages?
>> >
>> >Thanks.
>> >
>> >       [[alternative HTML version deleted]]
>> >
>> >______________________________________________
>> >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<http://www.r-project.org/posting-guide.html>
>> >and provide commented, minimal, self-contained, reproducible code.
>>
>>




More information about the R-help mailing list