Ultimate Linux Box (was RE: [R] optimal windows R machine)

[Peter Dalgaard BSA]

"Liaw, Andy" <andy_liaw at merck.com> writes:

> Well, maybe this is sort of related...
> 
> Eric Raymond has written an article for the Linux Journal on how to build an
> Ultimate Linux Box.  You can find the whole thing at
> 
> http://www2.linuxjournal.com/cgi-bin/frames.pl/articles/style/0014.html
> 
> The machine is said to compile the Linux 2.4.9 kernel in 1 min. 50 sec.  I
> wonder how long it might take to build R on this beast...

My guess is that it's about twice as fast as this beast (2xPentiumIII,866):

[Current R-devel]
291.64user 13.57system 4:01.62elapsed 126%CPU (0avgtext+0avgdata 0maxresident)k
0inputs+0outputs (586156major+348522minor)pagefaults 0swaps

The R builds spend quite a lot of time building the documentation,
and are locked to uniprocessor mode for the duration. This is because
it runs as a single perl script that processes all .Rd files in a
package. There's a good reason for that, but splitting things across a
couple of perl runs would speed things up considerably. The compile
part can be done (setting MAKE="make -j5") in 

165.89user 12.78system 1:46.26elapsed 168%CPU (0avgtext+0avgdata 0maxresident)k
0inputs+0outputs (634972major+360981minor)pagefaults 48swaps

(This machine is still running RedHat 6.2 with it's default dodgy
compiler, so make check doesn't run...)

-- 
   O__  ---- Peter Dalgaard             Blegdamsvej 3  
  c/ /'_ --- Dept. of Biostatistics     2200 Cph. N   
 (*) \(*) -- University of Copenhagen   Denmark      Ph: (+45) 35327918
~~~~~~~~~~ - (p.dalgaard at biostat.ku.dk)             FAX: (+45) 35327907
-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-
r-help mailing list -- Read http://www.ci.tuwien.ac.at/~hornik/R/R-FAQ.html
Send "info", "help", or "[un]subscribe"
(in the "body", not the subject !)  To: r-help-request at stat.math.ethz.ch
_._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._