high performance computing on graphics processing units: hgpu.org

The biomedical imagery, the numeric communications, the acoustic signal processing and many others digital signal processing applications (DSP) are present more and more everyday in the numeric world. They process growing data volume which is represented with more and more accuracy, and using complex algorithms with time constraints to satisfying. Consequently, a high requirement of computing power characterize them. To satisfy this need, it’s inevitable today to use parallel and heterogeneous architectures in order to speed-up the processing, where the best examples are the supercomputers like "Tianhe-2" and "Titan" of the ranking top500. These architectures with their multi-core nodes supported by many-core accelerators offer a good response to this problem, but they are still hard to program in order to make performance because of lot of things like synchronization, the memory management, the hardware specifications. In the present work, we propose a high level programming model to implement easily and efficiently digital signal processing applications on heterogeneous clusters.