pARMS is a library of parallel solvers for distributed sparse linear
systems of equations. It is based on a preconditioned Krylov subspace
approach, using a domain decomposition viewpoint. The plural in "Solvers"
is due to the fact that pARMS offers a large
selection of preconditioners for distributed sparse linear systems
and a few of the best known accelerators.
The basic methodology used relies on a Recursive
Multi-level ILU factorization wich allows to develop many of the standard
domain-decomposition type iterative solvers in a single framework.
For example, the standard Schwarz procedures are included as are
a number of Schur complement techniques. pARMS resulted from a team effort that spanned several years. It really began in around 1993/1994 with the development of its FORTRAN predecessor called PSPARSLIB [developer: Y. Saad]. In 1999-2000 Brian Suchomel developed the sequential version of ARMS. Zhongze Li developed the first version of the parallel ARMS in around 2000-2001. The current version of pARMS is the result of a substantial revision of this earlier code done in part by YS and to a bigger extent by Masha Sosonkina. Many people made other contributions to pARMS.
