Hypre
The Parallel High Performance Preconditioners (hypre) is a library of routines for scalable (parallel) solution of linear systems. The built-in BLOPEX package in addition allows solving eigenvalue problems. The main strength of Hypre is availability of high performance parallel multigrid preconditioners for both structured and unstructured grid problems.[1][2]
Stable release | 2.11.2
/ 2017/03/13 |
---|---|
Repository | |
Operating system | Linux, Unix |
Available in | C (main language), C++, FORTRAN |
Type | High-performance Parallel Software for linear systems and eigenvalue problems |
License | LGPL (version 2.1) |
Website | https://computing.llnl.gov/projects/hypre-scalable-linear-solvers-multigrid-methods |
Currently, Hypre supports only real double-precision arithmetic. Hypre uses the Message Passing Interface (MPI) standard for all message-passing communication. PETSc has an interface to call Hypre preconditioners.
Hypre is being developed and is supported by members of the Scalable Linear Solvers project within the Lawrence Livermore National Laboratory.
Features
hypre provides the following features:
- Parallel vectors and matrices, using several different interfaces
- Scalable parallel preconditioners
- Built-in BLOPEX
References
- Falgout, R.D.; Jones, J.E.; Yang, U.M. (2005). "Pursuing scalability for hypre's conceptual interfaces". ACM Transactions on Mathematical Software. 31 (3): 326–350. doi:10.1145/1089014.1089018. S2CID 19025723.
- Falgout, R.D.; Jones, J.E.; Yang, U.M. (2006). "The Design and Implementation of hypre, a Library of Parallel High Performance Preconditioners". In Bruaset, A. M.; Tveito, A. (eds.). Numerical Solution of Partial Differential Equations on Parallel Computers. Lecture Notes in Computational Science and Engineering. Vol. 51. Springer-Verlag. pp. 267–294. doi:10.1007/3-540-31619-1_8. ISBN 978-3-540-29076-6.