high performance computing on graphics processing units: hgpu.org

In our paper we present an abstract object oriented runtime system that helps to develop scientific application for new hererogenous architecture based on multi-node of multi-core processors enhanced with accelerator boards. Its architecture based on abstract concepts enables to follow hardware technology by extending them with new implementations modeling new hardware components, while limiting the impacts on existing application architecture or in the developpement of high level generative framework based on Domain specific language. We validate our approach with a multiscale algorithm to solve partial derivative equations that we have implemented with this runtime system and benchmarked on various heterogeneous hardware architecture.