Towards Ad-Hoc GPU Acceleration Of Parallel Eigensystem Computations

Michael T. Garba, Horacio Gonzalez-Velez
2011 ECMS 2011 Proceedings edited by: T. Burczynski, J. Kolodziej, A. Byrski, M. Carvalho  
This paper explores the early implementation of highperformance routines for the solution of multiple large Hermitian eigenvector and eigenvalue systems on a Graphics Processing Unit (GPU). We report a performance increase of up to two orders of magnitude over the original EISPACK routines with a NVIDIA Tesla C2050 GPU, potentially allowing an order of magnitude increase in the complexity or resolution of a neutron scattering modeling application.
doi:10.7148/2011-0534-0538 dblp:conf/ecms/GarbaG11 fatcat:w7aljmj2dzeproza5kq23np3zy