[petsc-users] Fastest solver option after factorization

[Hong Zhang]

Umut :
We rewrote petsc 'Forward backward solver' in Petsc-3.1 base on
@article{sz2010,
      author  = "B. Smith and H. Zhang",
      title   = "Sparse Triangular Solves for {ILU} Revisited: Data Layout
Crucial to Better Performance",
      journal = "International Journal of High Performance Computing
Applications",
      volume = 25,
      number = 4,
      pages = "386--391",
      year    = 2011
}
(note: this is for LU factorization, not symmetric Cholesky factorization).

We've never done comparison with other triangular solves.
It would be interesting to know.

Hong

Dear all,
>
> I am doing an investigation on the fastest 'Forward backward solver' among
> available sparse direct solvers. I am reading on comparison report, dated
> 2005, namely,
>
> A numerical evaluation of sparse direct solvers for the solution of large
> sparse, symmetric linear systems of equations
>
> N I M Gould, Y Hu, J A Scott
>
> And Pardiso seems to be the best among many solvers. Any further ideas on
> this topic?
>
> I guess this group is one the best lists to ask for opinions ;-)
>
> BR,
> Umut
>
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.mcs.anl.gov/pipermail/petsc-users/attachments/20120927/5c12d671/attachment.html>