[R] Way to make R idle for some time and try something again later

[Don MacQueen]

At 6:44 AM +0100 7/30/05, Prof Brian Ripley wrote:
>This depends on what else is going on.  My guess is that you are 
>running the Aqua GUI, and it is servicing the GUI which is taking 
>the time, not R itself.

Actually, no, I am not using the Aqua GUI. Not even a "framework" build:
    ../source/configure --enable-R-shlib --with-blas=-framework vecLib 
--with-lapack --enable-R-framework=no


>On all of Linux, Solaris and Windows (RGui or Rterm) Sys.sleep() 
>does use very close to zero resources at the beginning of a session, 
>but things may be different if e.g. tcltk widgets are in use.
>
>On Fri, 29 Jul 2005, Don MacQueen wrote:
>
>>I done something very similar -- have R watch a file, and whenever
>>new data is added to the file, read the new data from the file. In my
>>case, new data was arriving once per minute, so I needed to have R
>>wait about a minute before looking for new data.
>>
>>On my unix-based system, I found that if I usd
>
>I don't think your system IS `unix-based' (Unix is a trademark, and 
>MacOS X is based on a rather different kernel).  It is quite 
>possible that it is behaving differently from the POSIX description 
>of Unix system calls on which R is based for Unix-alikes.
>
>>       Sys.sleep( N )
>>then cpu usage immediately went up drastically. If the the system is
>>otherwise fairly idle, cpu usage goes up to nearly 100%. A cpu
>>monitor shows that R is using the cpu cycles.
>>
>>If I use instead
>>      system('sleep N')
>>cpu usage does not go up.
>
>Does that freeze the GUI?  It certainly freezes tcltk widgets on Unix.

I've never tried it while using tcltk widgets.

(and apologies for the delay; I've been away from the office for two weeks)

>
>>(where N is the number of seconds to sleep)
>>
>>>  version
>>          _
>>platform powerpc-apple-darwin7.9.0
>>arch     powerpc
>>os       darwin7.9.0
>>system   powerpc, darwin7.9.0
>>status
>>major    2
>>minor    1.1
>>year     2005
>>month    06
>>day      20
>>language R
>>
>>
>>At 12:13 PM -0700 7/29/05, Tae-Hoon Chung wrote:
>>>Hi, All;
>>>
>>>I have a question. In R, what is the best way to make R idle for a while and
>>>try something again later? For example, suppose there is an R job which
>>>accesses a file that may be shared with other active jobs. So when the file
>>>is being accessed by other job, your job will not be able to access the file
>>>and your job will crash because of that. To avoid this, you want your job to
>>>try to access the file repeatedly with some time interval, say every 10
>>>seconds or something like that. Which is the best way to do this in R?
>
>
>--
>Brian D. Ripley,                  ripley at stats.ox.ac.uk
>Professor of Applied Statistics,  http://www.stats.ox.ac.uk/~ripley/
>University of Oxford,             Tel:  +44 1865 272861 (self)
>1 South Parks Road,                     +44 1865 272866 (PA)
>Oxford OX1 3TG, UK                Fax:  +44 1865 272595


-- 
--------------------------------------
Don MacQueen
Environmental Protection Department
Lawrence Livermore National Laboratory
Livermore, CA, USA