[Rd] R on os390

[Strumila, John]

gday Thomas,

os390 uses 16/32 for int/long but the high order bit is not available, so
2gb is the best addressability.

The administrators of os390 tends to restrict virtual memory coz mainframe
RAM is historically expensive and paging is a very well understood/managed
resource.  Hence I don't know if I can physically grab 2Gb anyway.  On os390
this historically hasn't been a problem because it excels in I/O and
programs have been written with that in mind.  eg. SAS use of 'by' gets
around 'class' memory restrictions which means >2Gb files are handled
without much trouble.

I could only get nsize to about 200Mb (nsize=10,000K) on my redhat box.  Is
there something extra I need to do to increase this?

thanks,
John Strumila
 

-----Original Message-----
From: Thomas Lumley [mailto:thomas@biostat.washington.edu]
Sent: Saturday, 19 August 2000 7:46
To: Strumila, John
Cc: r-devel@stat.math.ethz.ch
Subject: Re: [Rd] R on os390


On Fri, 18 Aug 2000, Strumila, John wrote:

> G'day R friends,
> 
> I didn't get any replies on the main list so I thought I'd try with the
> experts.
> 
> I was wondering if anyone's ported R to os390.  If so, are the vsize and
> nsize limits the same as other platforms?

I'm not aware of a port.

The vsize and nsize limits would take some surgery to fix, but they
already allow 2Gb of heap and 2Gb of cons cells, which is quite a lot.

I would like R to be able to use large memory, since it's becoming more
widely available, but quite a few changes are probably needed.

What size are int, long and pointer types on os390?


> I could really annoy those SAS guys then.

:)

	-thomas

Thomas Lumley
Assistant Professor, Biostatistics
University of Washington, Seattle
-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-
r-devel 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-devel-request@stat.math.ethz.ch
_._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._