High-Precision Anchored Accumulators for Reproducible Floating-Point Summation
High-Precision Anchored Accumulators for Reproducible Floating-Point Summation
复制标题
用于可重复浮点求和的高精度锚定累加器
DOI：
--
复制
发表时间：
2017
期刊：
影响因子：
--
通讯作者：
C. Hinds
中科院分区：
文献类型：
--
作者：
D. Lutz;C. Hinds
本文介绍了一种新的数据类型，允许在程序员可选择的范围内重复累加浮点数 (FP)。新的数据类型比现有的 FP 格式具有更大的有效数和更小的范围，并且具有更好的算术和计算特性。特别是，它是关联的、可并行的、可再现的和正确的。对于能够解决大多数问题的适度范围，它的速度也快得多：在单个 256 位 SIMD 实现上快 3 到 12 倍。本文还描述了支持所提出的数据类型的新指令和相关数据路径，并讨论了如何使用所提出的方法来实现最近发布的可再现 FP 求和的软件算法。
This paper introduces a new datatype that allows reproducible accumulation of floating-point (FP) numbers in a programmer-selectable range. The new datatype has a larger significand and a smaller range than existing FP formats and has much better arithmetic and computational properties. In particular, it is associative, parallelizable, reproducible and correct. For the modest ranges that will accommodate most problems, it is also much faster: 3 to 12 times faster on a single 256-bit SIMD implementation. The paper also describes a new instruction and associated datapath that support the proposed datatype, and discusses how a recently published software algorithm for reproducible FP summation could be implemented using the proposed approach.