A new memory mapping mechanism for GPGPUs' stencil computation
DOI10.1007/s00607-014-0434-5zbMath1321.65135MaRDI QIDQ2516110
Publication date: 10 August 2015
Published in: Computing (Search for Journal in Brave)
Full work available at URL: https://doi.org/10.1007/s00607-014-0434-5
numerical example; heat equation; finite difference method; performance evaluation; shared memory; GPU; stencil computation; memory bandwidth; control flow divergence; coalesced memory; ghost zone; memory access traffic; memory mapping; software prefetching
35K05: Heat equation
65M06: Finite difference methods for initial value and initial-boundary value problems involving PDEs
65M50: Mesh generation, refinement, and adaptive methods for the numerical solution of initial value and initial-boundary value problems involving PDEs
65Y20: Complexity and performance of numerical algorithms
Uses Software
Cites Work