...
首页> 外文期刊>IEEE Transactions on Computers >Single Precision Logarithm and Exponential Architectures for Hard Floating-Point Enabled FPGAs
【24h】

Single Precision Logarithm and Exponential Architectures for Hard Floating-Point Enabled FPGAs

机译:硬浮点FPGA的单精度对数和指数架构

获取原文
获取原文并翻译 | 示例
           

摘要

In this article we present a novel method for implementing floating point (FP) elementary functions using the new FP single precision addition and multiplication features of the Arria 10 and Stratix 10 DSP Block architecture. Our application examples are $log (x)$ and $exp (x)$ , two of the most commonly required functions for emerging datacenter and computing FPGA targets. We explain why the combination of new FPGA technology, and at the same time, a massive increase in computing performance requirement, fuels the need for this work. We show a comprehensive error analysis, and discuss various implementation trade-offs that demonstrate that the hard FP (HFP) Blocks, in conjunction with the traditional flexibility and connectivity of the FPGA, can provide a robust and high performance solution. The architectures presented in this work meet OpenCL accuracy requirements. Our methods map extensively to embedded structures, and therefore result in significant reduction in logic resources and routing stress compared to current methods. The methods allow leveraging the routing architectures introduced in the Stratix 10 device which results in high-function performance.
机译:在本文中,我们介绍了一种新颖的方法,该方法使用Arria 10和Stratix 10 DSP模块架构的新FP单精度加法和乘法功能来实现浮点(FP)基本功能。我们的应用示例为 $ log(x)$ $ exp(x)$ ,这是新兴数据中心和计算FPGA目标最常用的两个功能。我们解释了为什么结合使用新的FPGA技术,同时又大大提高了计算性能要求,从而推动了这项工作的需要。我们展示了全面的错误分析,并讨论了各种实现折衷方案,这些方案证明了硬FP(HFP)模块与FPGA的传统灵活性和连通性相结合,可以提供一个健壮的高性能解决方案。本工作中介绍的体系结构满足OpenCL准确性要求。我们的方法广泛地映射到嵌入式结构,因此与当前方法相比,可显着减少逻辑资源和布线压力。这些方法允许利用Stratix 10器件中引入的路由架构,从而实现高性能。

著录项

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱:kefu@zhangqiaokeyan.com

京公网安备:11010802029741号 ICP备案号:京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有
  • 客服微信

  • 服务号