1lis(3) Introduction to Library Functions lis(3)
2
3
4
6 lis - library of iterative solvers for linear systems
7
8
10 [see Lis User Guide for full description]
11
12
14 Lis (Library of Iterative Solvers for linear systems, pronounced [lis])
15 is a parallel software library for solving discretized linear equations
16 and eigenvalue problems that arise in the numerical solution of partial
17 differential equations using iterative methods.
18
19
20 Lis provides facilities for:
21
22 * Automatic program configuration
23
24 * NUMA aware hybrid implementation with MPI and OpenMP
25
26 * Exchangeable dense and sparse matrix storage formats
27
28 * Basic linear algebra operations for dense and sparse matrices
29
30 * Parallel iterative methods for linear equations and eigenvalue prob‐
31 lems
32
33 * Parallel preconditioners for iterative methods
34
35 * Quadruple precision floating point operations
36
37 * Performance analysis
38
39 * Command-line interface to solvers and benchmarks
40
41
43 The installation of Lis requires a C compiler. The Fortran interface
44 requires a Fortran compiler, and the algebraic multigrid preconditioner
45 requires a Fortran 90 compiler. For parallel computing environments, an
46 OpenMP or an MPI-1 library is used. Both the Harwell-Boeing and Matrix
47 Market formats are supported to import and export user data.
48
49
51 * Akira Nishida (2010). "Experience in Developing an Open Source Scal‐
52 able Software Infrastructure in Japan". Computational Science and Its
53 Applications - ICCSA 2010. Lecture Notes in Computer Science 6017.
54 Springer. pp. 87-98. doi:10.1007/978-3-642-12165-4_36. ISBN
55 3-642-12164-0.
56
57 * Hisashi Kotakemori, Hidehiko Hasegawa, Tamito Kajiyama, Akira Nukada,
58 Reiji Suda, and Akira Nishida (2008). "Performance Evaluation of Paral‐
59 lel Sparse Matrix-Vector Products on SGI Altix 3700". OpenMP Shared
60 Memory Parallel Programming. Lecture Notes in Computer Science 4315.
61 Springer. pp. 153-163. doi:10.1007/978-3-540-68555-5_13. ISBN
62 3-540-68554-5.
63
64 * Hisashi Kotakemori, Hidehiko Hasegawa, and Akira Nishida (2005).
65 "Performance Evaluation of a Parallel Iterative Method Library using
66 OpenMP". Proceedings of the 8th International Conference on High Per‐
67 formance Computing in Asia Pacific Region (HPC Asia 2005). Beijing:
68 IEEE. pp. 432-436. doi:10.1109/HPCASIA.2005.74. ISBN 0-7695-2486-9.
69
70 * Akihiro Fujii, Akira Nishida, and Yoshio Oyanagi (2005). "Evaluation
71 of Parallel Aggregate Creation Orders : Smoothed Aggregation Algebraic
72 Multigrid Method". High Performance Computational Science and Engineer‐
73 ing. Springer. pp. 99-122. doi:10.1007/0-387-24049-7_6. ISBN
74 1-4419-3684-X.
75
76
78 lsolve(1), esolve(1), hpcg_kernel(1), hpcg_spmvtest(1), spmvtest1(1),
79 spmvtest2(1), spmvtest2b(1), spmvtest3(1), spmvtest3b(1), spmvtest4(1),
80 spmvtest5(1)
81
82 http://www.ssisc.org/lis/
83 http://math.nist.gov/MatrixMarket/
84
85
86
87
88Man Page 25 Oct 2016 lis(3)