Sciweavers

PROCEDIA
2010
103views more  PROCEDIA 2010»
12 years 11 months ago
Towards generating optimised finite element solvers for GPUs from high-level specifications
We argue that producing maintainable high-performance implementations of finite element methods for multiple targets requires that they are written using a high-level domain-speci...
Graham R. Markall, David A. Ham, Paul H. J. Kelly