Lis: a Library of Iterative Solvers for Linear Systems

Download | Publications | Contact

Introduction

Lis, a Library of Iterative Solvers for linear systems, is a scalable parallel library for solving linear equations and standard eigenvalue problems with real sparse matrices using iterative methods.
Installation of Lis requires a C compiler. To use the Fortran interface, a FORTRAN 77 compiler is required. To use the algebraic multigrid preconditioner, a Fortran 90 compiler is required. OpenMP and MPI-1 are used in parallel computing environments.
Lis has been tested on the following platforms:

• Version 1.2.35 (July 29, 2010) New!
  • lis-1.2.35.tar.gz (Source Code)
  • lis-manual-en.pdf (Users Manual)
• Version 1.1.2 (September 24, 2007)
  • lis-1.1.2.tar.gz (Source Code)
  • lis-manual-1.1.2_en.pdf (Users Manual)
  • lis-1.1.1-win32.zip (Binary Package for Windows)
• ChangeLog

• A. Nishida. Experience in Developing an Open Source Scalable Software Infrastructure in Japan, In Proceedings of Computational Science and its Applications (ICCSA 2010), Part II, Lecture Notes in Computer Science 6017, pp. 87-98, Springer, 2010.
• A. Nishida. Performance Optimization of Sparse Matrix-Vector Product on Multiple Architectures, IPSJ SIG Notes, Vol.2009-ARC-186 No.7, pp.1-8, 2009 (in Japanese).
• A. Nishida. Development of General Purpose Numerical Software Infrastructure for Large Scale Scientific Computing, In Annual Report of the Earth Simulator Center, April 2008 - September 2008, The Earth Simulator Center, Japan Agency for Marine-Earth Science and Technology, pp.129-133, 2009. (paper)
• A. Nishida. Algorithm of Parallle Nonlinear Conjugate Gradient Method and its Performance Evaluation. IPSJ SIG Technical Report, 2009-HPC-121(36), pp.1-6, 2009 (in Japanese).
• A. Nishida, Development of General Purpose Numerical Software Infrastructure for Large Scale Scientific Computing, In Annual Report of the Earth Simulator Center, April 2007 - March 2008, The Earth Simulator Center, Japan Agency for Marine-Earth Science and Technology, pp.137-145, 2008. (paper)
• A. Nishida. Implementation and Evaluation of Eigensolvers using Parallel Linear Algebra Library Lis, IPSJ SIG Notes, 2008(74), pp.145-150, 2008 (in Japanese).
• H. Kotakemori, A. Fujii, H. Hasegawa, and A. Nishida. Implementation of Fast Quad Precision Operation and Acceleration with SSE2 for Iterative Solver Library. In IPSJ Transactions on Advanced Computing Systems, Vol. 1, No. 1, pp.75-84, 2008 (in Japanese).
• H. Kotakemori, A. Fujii, H. Hasegawa and A. Nishida. Stabilization of Krylov subspace methods using fast quadruple-precision operation (in Japanese). In Proceedings of the 12th Annual Conference of the Japan Society for Computational Engineering and Science, pp. 631-634, 2007. (Best Paper Award, paper, slides).
• H. Kotakemori, A. Fujii, H. Hasegawa and A. Nishida. A proposal of mixture type iterative method of double precision and quadruple precision (in Japanese). In Proceedings of High Performance Computing Symposium 2007 (HPCS2007), pp. 9-16, IPSJ, 2007. (paper, slides)
• A. Nishida, H. Kotakemori, T. Kajiyama, and A. Nukada. Scalable Software Infrastructure Project. In Proceedings of the IEEE/ACM SC06 Conference, poster, November 11-17, 2006 (poster).
• H. Kotakemori, A. Fujii, H. Hasegawa and A. Nishida. Fast Quad Precision Iterative Solver Library Lis using SSE2 (in Japanese). IPSJ SIG Technical Report, 2006-HPC-108, pp. 7-12, 2006. (paper, slides)
• H. Kotakemori, H. Hasegawa, and A. Nishida, Performance Evaluation of a Parallel Iterative Method Library using OpenMP. In Proceedings of the 8th International Conference on High Performance Computing in Asia Pacific Region (HPC Asia 2005), pp.432-436, IEEE, 2005. (abstract/slides)
• H. Kotakemori, H. Hasegawa, T. Kajiyama, A. Nukada, R. Suda, and A. Nishida. Performance Evaluation of Parallel Sparse Matrix-Vector Products on SGI Altix 3700. In Proceedings of the First International Workshop on OpenMP (IWOMP2005), Lecture Notes in Computer Science 4315, pp.153-163, Springer, 2008. (abstract/slides)
• A. Fujii, A. Nishida, and Y. Oyanagi, Evaluation of Parallel Aggregate Creation Orders : Smoothed Aggregation Algebraic Multigrid Method, In High Performance Computational Science And Engineering, pp.99-122, Springer, 2005.
• A. Nishida. A Short Survey of Applications and Evaluations of Preconditioned Conjugate Gradient Method for Large Scale Eigenvalue Problems, In Proceedings of the 2003 annual conference of the Japan Society for Industrial and Applied Mathematics, Kyoto University, pp.326-327, 2003 (in Japanese).

Bug reports, requests and comments are highly appreciated. Please feel free to send e-mail to the following address:

The SSI Project