Towards ad-hoc GPU acceleration of parallel eigensystem computations
IDEAS Research Institute, Robert Gordon University, Aberdeen, UK
25th European Conference on Modelling and Simulation (ECMS 2011), 2011
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.
January 13, 2012 by hgpu
