HUTI: Framework for Iterative Solvers

Authors:
Harri Hakula;Juha Ruokolainen;Jouni Malinen
Affiliations:
-;-;-
Venue:
PARA '02 Proceedings of the 6th International Conference on Applied Parallel Computing Advanced Scientific Computing
Year:
2002

Citing 6
Cited 0

A transpose-free quasi-minimal residual algorithm for non-Hermitian linear systems

SIAM Journal on Scientific Computing
An implementation of the QMR method based on coupled two-term recurrences

SIAM Journal on Scientific Computing
Using MPI: portable parallel programming with the message-passing interface

Using MPI: portable parallel programming with the message-passing interface
The Parallel Iterative Methods (PIM) package for the solution of systems of linear equations on parallel computers

Applied Numerical Mathematics - Special issue on massively parallel computing and applications
QMRPACK: a package of QMR algorithms

ACM Transactions on Mathematical Software (TOMS)
Iterative Methods for Sparse Linear Systems

Iterative Methods for Sparse Linear Systems

Quantified Score

Hi-index	0.00

Visualization

Abstract

HUTI is a framework for development of libraries of iterative methods, especially the Krylov methods. For every algorithm the same implementation is used for all memory models and architectures, including the parallel systems. This leads to significant benefits in maintenance and debugging. The callback function approach has been employed extensively. This makes it possible to parallelize HUTI-based algorithms simply by rewriting the necessary callbacks. This flexibility comes with a price, however, for the responsibility for selecting the appropriate matrix data structures has been delegated to the user. Thus, HUTI itself cannot be used to solve any systems directly but it can easily be embedded into domain-specific solvers.