MPI-CUDA parallel linear solvers for block-tridiagonal matrices in the context of SLEPc&apos

[EN] We consider the computation of a few eigenpairs of a generalized eigenvalue problem Ax = lambda Bx with block-tridiagonal matrices, not necessarily symmetric, in the context of Krylov methods. In this kind of computation, it is often necessary to solve a linear system of equations in each itera...

Descripción completa

Detalles Bibliográficos
Autores: Lamas Daviña, Alejandro, Jose E. Roman|||0000-0003-1144-6772
Tipo de recurso: artículo
Fecha de publicación:2018
País:España
Institución:Universitat Politècnica de València (UPV)
Repositorio:RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Idioma:inglés
OAI Identifier:oai:riunet.upv.es:10251/125676
Acceso en línea:https://riunet.upv.es/handle/10251/125676
Access Level:acceso abierto
Palabra clave:MPI
GPU computing
Eigenvalue computation
Block-tridiagonal linear solvers
CIENCIAS DE LA COMPUTACION E INTELIGENCIA ARTIFICIAL
Descripción
Sumario:[EN] We consider the computation of a few eigenpairs of a generalized eigenvalue problem Ax = lambda Bx with block-tridiagonal matrices, not necessarily symmetric, in the context of Krylov methods. In this kind of computation, it is often necessary to solve a linear system of equations in each iteration of the eigensolver, for instance when B is not the identity matrix or when computing interior eigenvalues with the shift-and-invert spectral transformation. In this work, we aim to compare different direct linear solvers that can exploit the block-tridiagonal structure. Block cyclic reduction and the Spike algorithm are considered. A parallel implementation based on MPI is developed in the context of the SLEPc library. The use of GPU devices to accelerate local computations shows to be competitive for large block sizes.