Re: optimizing vector integrations

["Robert G. Brown" <rgb at phy dot duke dot edu>]

On Fri, 31 Jan 2003, [ISO-8859-15] Achim Gädke wrote:

> Hi folks!
> 
> I have to integrate a vector-function, which returns its values once. 
> I'd like to use an adaptive algorithm, which profits of this behavior, 
> instead of ignoring one value.

GSL ODE solvers handle vectors, are now adaptive, and can be used to
"plain old" integrate, actually.  The only difference betweeen
integrating a vector of differentials (if that is what you mean by
"integrating a vector function) and solving coupled ODEs is that in the
latter, the differentials are coupled.

As for WHICH ODE solver to use, it is hard to say.  I'd probably start
with good old runga-kutta (adequate for most non-stiff problems and
likely more than adequate for ordinary integration).  There are other
algorithms, though, that will probably be faster, so experiment.  One
nice thing about all of the GSL tools is that they seem to be designed
as algorithm harnesses -- write your code once, and then change
algorithms with a single #define, variable assignment, or even
shell-level environment variable.  This really makes experimentation
easy.

   rgb

> 
> Any suggestions?
> 
> Achim
> 

-- 
Robert G. Brown	                       http://www.phy.duke.edu/~rgb/
Duke University Dept. of Physics, Box 90305
Durham, N.C. 27708-0305
Phone: 1-919-660-2567  Fax: 919-660-2525     email:rgb@phy.duke.edu