Quantifying the Interference Caused by Subnormal Floating-Point Values
Quantifying the Interference Caused by Subnormal Floating-Point Values
复制标题
量化非正规浮点值引起的干扰
DOI：
--
复制
发表时间：
2006
期刊：
影响因子：
--
通讯作者：
N. Goodwin
中科院分区：
文献类型：
--
作者：
Isaac Dooley;L. Kalé;N. Goodwin
当使用或产生次正常（也称为非规范化或非规范化）浮点值时，普通浮点操作会很慢。本文描述了如何通过提供一个简单的微基准来量化低于正常值引起的放缓程度。以前的工作已经表明，由次法线引起的减速对于某些真实的世界应用是重要的，并且次法线的出现不应该仅仅被视为例外情况[1]。本文提出了一个微基准测试，然后分析了在各种现代微体系结构上运行C和Java中的微基准测试的结果。结果表明，所有常见的现代微架构都容易受到这种类型的不希望的减速。
Ordinary floating point operations are slow when subnormal(also called denormalized or unnormalized) floating point values are used or produced. This paper describes how to quantify the extent of the slowdown caused by subnormal values by providing a simple microbenchmark. Previous work has shown that the slowdown caused by subnormals is significant for some real world applications and that the occurrence of subnormals should not just be treated as an exceptional case[1]. This paper both proposes a micro-benchmark and then analyzes the results for running the proposed micro-benchmark in both C and Java on a variety of modern microarchitectures. The results presented show that all common modern micro-architectures are susceptible to this type of undesirable slowdown.