FMMTL: FMM Template Library A Generalized Framework for Kernel Matrices
In response to two decades of development in structured dense matrix algorithms and a vast number of research codes, we present designs and progress towards a codebase that is abstracted over the primary domains of research. In the domain of mathematics, this includes the development of interaction kernels and their low-rank expansions. In the domain of high performance computing, this includes the optimized construction, traversal, and scheduling algorithms for the appropriate operations. We present a versatile system that can encompass the design decisions made over a decade of research while providing an abstracted, intuitive, and usable front-end that can integrated into existing linear algebra libraries.
Unable to display preview. Download preview PDF.
- 1.E. Agullo, B. Bramas, O. Coulaud, E. Darve, M. Messner, T. Takahashi, Pipelining the Fast Multipole Method over a Runtime System. Research Report RR-7981, INRIA, May 2012Google Scholar
- 8.H. Ltaief, R. Yokota, Data-driven execution of fast multipole methods. Concurr. Comput.: Pract. Experience, pp. n/a–n/a (2013)Google Scholar