No abstract

No abstract

Node level heterogeneous architectures have become attractive in recent years for several reasons: Compared to traditional symmetric CPUs, they offer high performance for real applications, and can be energy and/or cost efficient. In this paper, we give an overview of the state-of-the-art in heterogeneous computing, focusing on some common architectures: The NVidia and the ATI graphics processing units (GPUs), the Cell Broadband Engine Architecture (CBEA), the ClearSpeed processor, the field programmable gate array (FPGA) accelerator solutions from Maxeler MaxNodes (MAX), the SGI systems (RASC), and the Convey Hybrid-Core Computer. We present a review of the hardware, available software tools for each solution, a quantitative and a qualitative comparison of the architectures, and we give our view on the future of heterogeneous computing.

No abstract

No abstract