Harnessing FPGAs potential with OpenCL

The work presented deals with the evaluation of F-PGAs resurgence for hardware and software acceleration. We focus our attention on the tools developed by FPGAs manufacturers, in particular the Intel FPGA S-DK for OpenCL, that promises a new level of hardware abstraction from the developer's perspective, allowing a software-like programming of FPGAs. Our first contribution is to propose an accurate memory benchmark, and we follow with an evaluation of different custom OpenCL implementations of one use case : the computed tomogra-phy. With some clues on memory fetching and coalescing, we then further tune designs to improve performance. Finally , a comparison is made with GPU implementations, and a preliminary conclusion is drawn on FPGAs future in the semiconductor realm. This presentation include but is not limited to results presented in [3].