Hello!

I'm new to GSL and want to know what kind of linear solver is best
suited for my problem.

I have linear systems of type Ax = b with quadratic A. The dimension is
moderate in size, somewhat between 20 and 100, but I need to solve a
large amount systems, about 50,000, so efficiency is a matter. Since
matrices are moderate in size, CPU efficiency is probably more important
than memory efficiency.

In the GSL documentation I found this example:
http://www.gnu.org/software/gsl/manual/html_node/Linear-Algebra-Examples.html
which solves a system using LU decomposition of A.

Another method I found is:
http://www.gnu.org/software/gsl/manual/html_node/Householder-solver-for-linear-systems.html

I want to know if there are other methods to solve linear systems and
which method is best suited for my problem regarding efficiency. The
in-place Householder-method seems to be *very* memory efficient, but is
it as fast as the other Householder method? I can't find any hints on that.

Thanks,
Simon


_______________________________________________
Help-gsl mailing list
[email protected]
http://lists.gnu.org/mailman/listinfo/help-gsl

Reply via email to