[STATSGRASS] gstat limitation?

[Roger Bivand]

On Fri, 28 Jul 2006, Carlos "Guâno" Grohmann wrote:

> "Killed" is the only message I get. It starts running the kriging and
> then quits R and say "killed".

Is it possible to establish what is generating the message? Can you run R 
-d gdb for example to backtrack? 

Since the max. distance is restricted, can you tile the input data and 
output data with overlaps to try to find out a maximum acceptable region? 
Probably we need Edzer's input, and I think he is offline now?

Roger

> 
> Carlos
> 
> 
> On 7/28/06, Roger Bivand <Roger.Bivand at nhh.no> wrote:
> > On Thu, 27 Jul 2006, Carlos "Guâno" Grohmann wrote:
> >
> > > I am running it in R. it get killed in the kriging step. I guess its
> > > abot system memory.
> >
> > What does "killed" mean? Specifically, what is the error message? Is it a
> > malfunction?
> >
> > If I understand, you have about 300000 data points and want to predict to
> > about 2.5 million, using a very small maximum distance. I would have
> > looked at subsetting, if there is no other way round.
> >
> > Roger
> >
> > >
> > > Carlos
> > >
> > >
> > > On 7/27/06, Roger Bivand <Roger.Bivand at nhh.no> wrote:
> > > > On Thu, 27 Jul 2006, Carlos "Guâno" Grohmann wrote:
> > > >
> > > > > Hello
> > > > > I was wondering if gstat has any limitation about file size it can
> > > > > process. I want to resample a DEM of a large area (about 50x50 km, 30m
> > > > > res.) with kriging, but it get killed.
> > > >
> > > > That sounds painful, but it doesn't tell much. Without your verbatim
> > > > commands, it is difficult to tell - are you running gstat standalone, or
> > > > in R? At what step in the modelling process does the problem arise?
> > > >
> > > > Roger
> > > >
> > > > >
> > > > > Carlos
> > > > >
> > > > >
> > > >
> > > > --
> > > > Roger Bivand
> > > > Economic Geography Section, Department of Economics, Norwegian School of
> > > > Economics and Business Administration, Helleveien 30, N-5045 Bergen,
> > > > Norway. voice: +47 55 95 93 55; fax +47 55 95 95 43
> > > > e-mail: Roger.Bivand at nhh.no
> > > >
> > > >
> > >
> > >
> > >
> >
> > --
> > Roger Bivand
> > Economic Geography Section, Department of Economics, Norwegian School of
> > Economics and Business Administration, Helleveien 30, N-5045 Bergen,
> > Norway. voice: +47 55 95 93 55; fax +47 55 95 95 43
> > e-mail: Roger.Bivand at nhh.no
> >
> >
> >
> 
> 
> 

-- 
Roger Bivand
Economic Geography Section, Department of Economics, Norwegian School of
Economics and Business Administration, Helleveien 30, N-5045 Bergen,
Norway. voice: +47 55 95 93 55; fax +47 55 95 95 43
e-mail: Roger.Bivand at nhh.no