[R] Choosing hardware

[Patrick Connolly]

On Mon, 04-Feb-2008 at 10:42AM -0700, Todd Remund wrote:

|> 

|> I'm ordering a new computer to increase my ability to handle large
|> data sets.  I?ve tried the dual core type and also the dual
|> processor with dual cores each, and have not been satisfied.  This
|> seems to agree with all the other postings on the help list.  I
|> don?t want to do any simulations, I just want to deal with a large
|> data set and do things such as resampling time series data from
|> 50,000 samples per second to 2,000 samples per second.  And compute
|> other things that are not easily done on my PC.  What type of
|> hardware do I want?  I am leaning toward a single processor, since
|> I have had a slow down with the dual core processors.  It seems to
|> me that the threading and clustering will not really help my
|> situation.  Does anyone have any suggestions?

Sounds as though it's a memory limitation.  If that's the case, you
might do best with a 64bit machine and a decent amount of memory.  If
you're using Linux, it's easy since 64 bit has been working for years.
If you have to use Windows (you didn't say), you might have to wait a
bit longer.

best


-- 
~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.   
   ___    Patrick Connolly   
 {~._.~}          		 Great minds discuss ideas    
 _( Y )_  	  	        Middle minds discuss events 
(:_~*~_:) 	       		 Small minds discuss people  
 (_)-(_)  	                           ..... Anon
	  
~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.