arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2307.00124math.NAcs.NA
收藏

使用块浮点算术的多重网格方法

Multigrid Methods using Block Floating Point Arithmetic

Nils Kohl, Stephen F. McCormick, Rasmus Tamstorf

AI总结:

本文探讨了块浮点(BFP)算术在混合精度和渐进精度多重网格方法中的应用,通过提供类BLAS例程算法和基于饱和与截断的启发式方法,实现了节能且硬件高效的整数算术来求解线性椭圆偏微分方程,同时达到离散误差精度。

摘要:

块浮点(BFP)算术目前正重新引起人们的兴趣,因为与标准浮点算术相比,它需要更少的功耗、更少的芯片面积,并且在硬件中实现起来不那么复杂。本文探讨了BFP在混合精度和渐进精度多重网格方法中的应用,使得能够以节能且硬件高效的整数算术来求解线性椭圆偏微分方程(PDE)。虽然BFP算术的大多数现有应用倾向于使用较小的块大小,但这里选择的块大小是最大的,使得矩阵和向量的所有条目共享一个单一的指数。这有时也被称为缩放定点格式。我们为BFP算术的类BLAS例程提供了算法,确保在指定精度内的精确向量-向量和矩阵-向量计算。使用这些算法,我们研究了达到离散误差精度所需的渐近精度要求。我们证明了一些计算可以使用低至4位整数来执行,而达到特定目标精度所需的位数与标准浮点算术相似。最后,我们提出了一种基于饱和和截断的BFP算术全多重网格启发式方法,该方法仍然能够达到离散误差精度,而无需对中间结果进行昂贵的归一化步骤。

英文摘要:

Block Floating Point (BFP) arithmetic is currently seeing a resurgence in interest because it requires less power, less chip area, and is less complicated to implement in hardware than standard floating point arithmetic. This paper explores the application of BFP to mixed- and progressive-precision multigrid methods, enabling the solution of linear elliptic partial differential equations (PDEs) in energy- and hardware-efficient integer arithmetic. While most existing applications of BFP arithmetic tend to use small block sizes, the block size here is chosen to be maximal such that matrices and vectors share a single exponent for all entries. This is sometimes also referred to as a scaled fixed-point format. We provide algorithms for BLAS-like routines for BFP arithmetic that ensure exact vector-vector and matrix-vector computations up to a specified precision. Using these algorithms, we study the asymptotic precision requirements to achieve discretization-error-accuracy. We demonstrate that some computations can be performed using as little as 4-bit integers, while the number of bits required to attain a certain target accuracy is similar to that of standard floating point arithmetic. Finally, we present a heuristic for full multigrid in BFP arithmetic based on saturation and truncation that still achieves discretization-error-accuracy without the need for expensive normalization steps of intermediate results.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑