Configurable Computing: The Road Ahead
Configurable Computing: The Road Ahead
复制标题
可配置计算：未来之路
DOI：
--
复制
发表时间：
1997
期刊：
影响因子：
--
通讯作者：
B. Hutchings
中科院分区：
文献类型：
--
作者：
W. Mangione;B. Hutchings
至少从1989年开始，基于FPGA的计算平台已经证明了在一系列应用中实现极高性能的潜力：图像过滤，卷积，形态，特征提取和对象跟踪。高性能的潜力是真实的，对于许多应用程序，研究人员开发了实现性能的原型系统，该系统比传统方法高的数量级[A93]。但是，事实证明，意识到实验室之外的这种潜力非常困难，因为这些系统依赖于操纵低水平的抽象（例如数字电路），因此需要高技能的开发人员。本文的目的是讨论这项重要技术的过去，现在和未来，并确定哪些方法奏效了，这些方法目前表现出希望，哪些方法可能会在未来成功。
Since at least 1989, FPGA-based computing platforms have demonstrated the potential for achieving extremely high performance for a range of applications: image filtering, convolution, morphology, feature extraction, and object tracking. The potential for high performance is real, and for a number of applications researchers have developed prototype systems that achieve performance that is an order of magnitude higher than conventional approaches [A93]. However, realizing this potential outside of the laboratory has proven extremely difficult because these systems rely on manipulating low level abstractions (e.g. digital circuits) and thus require highly skilled developers. The goal of this paper is to discuss the past, present and future of this important technology and to determine which approaches have worked, which are currently showing promise, and which will likely prove successful in the future.