作者: Maciej Wielgosz , Ernest Jamro , Kazimierz Wiatr
DOI: 10.1109/ICSES.2008.4673367
关键词:
摘要: This paper presents results of the tests performed to determine high speed calculations capabilities SGI RASC platform. Different data transfer modes and memory management approaches were examined choose most effective combination Host adjustments. Obtained measurements revealed that Direct I/O mode together with DMA provides highest throughput between slice. Nevertheless, for some applications multi-buffering may be more suitable in terms concurrent FPGA algorithm execution capability.