SPIKE: A parallel environment for solving banded linear systems

Abstract The hybrid banded linear solver SPIKE is proposed as a parallel environment for solving banded systems that are either dense or sparse within the band. The SPIKE algorithm is a domain decomposition technique that allows performing independent calculations on each subdomain or partition of the original linear system. The interface problem leads to a reduced linear system of much smaller size than that of the original system. Three different members of the SPIKE family are described. Each handles the reduced system in a different way depending on the characteristics of the system and the architecture of the high-end parallel computing platform. Numerical experiments are presented that demonstrate the effectiveness of our parallel scheme. Comparison with the corresponding algorithms of ScaLAPACK are also provided for those banded systems that are dense within the band. A SPIKE scheme with multi-level parallelism is also introduced for solving large banded systems that are sparse within the band.

[1]  Patrick Amestoy,et al.  A Fully Asynchronous Multifrontal Solver Using Distributed Dynamic Scheduling , 2001, SIAM J. Matrix Anal. Appl..

[2]  James Demmel,et al.  SuperLU_DIST: A scalable distributed-memory sparse direct solver for unsymmetric linear systems , 2003, TOMS.

[3]  Ahmed H. Sameh,et al.  A parallel hybrid banded system solver: the SPIKE algorithm , 2006, Parallel Comput..

[4]  Jack J. Dongarra,et al.  A Comparison of Parallel Solvers for Diagonally Dominant and General Narrow-Banded Linear Systems , 1999, Scalable Comput. Pract. Exp..

[5]  Michael Hatzopoulos,et al.  A parallel linear system solver , 1979 .

[6]  William Jalby,et al.  Parallel Algorithms on the CEDAR System , 1986, CONPAR.

[7]  William F. Moss,et al.  Decay rates for inverses of band matrices , 1984 .

[8]  Michael W. Berry,et al.  Multiprocessor Schemes for Solving Block Tridiagonal Linear Systems , 1988 .

[9]  Duncan H. Lawrie,et al.  The computation and communication complexity of a parallel banded system solver , 1984, TOMS.

[10]  Jack Dongarra,et al.  A comparison of parallel solvers for diagonally dominant and general narrow-banded linear systems , 2001 .

[11]  David J. Kuck,et al.  On Stable Parallel Linear System Solvers , 1978, JACM.

[12]  Eric Polizzi,et al.  Self-consistent three-dimensional models for quantum ballistic transport in open systems , 2002 .

[13]  Duncan H. Lawrie,et al.  High Speed Computer and Algorithm Organization , 1977 .

[14]  Tayfun E. Tezduyar,et al.  Enhanced‐discretization successive update method (EDSUM) , 2005 .

[15]  Tayfun E. Tezduyar,et al.  Enhanced-approximation linear solution technique (EALST) , 2004 .

[16]  Jack Dongarra,et al.  ScaLAPACK user's guide , 1997 .

[17]  Ahmed Sameh,et al.  Parallel direct linear system solvers -- a survey☆ , 1977 .

[18]  Tayfun E. Tezduyar,et al.  Automatic mesh update with the solid-extension mesh moving technique , 2004 .