计算机科学 ›› 2026, Vol. 53 ›› Issue (2): 124-132.doi: 10.11896/jsjkx.250200014

• 计算机体系结构 • 上一篇    下一篇

基于网格化粒子群搜索算法的最大浮点误差并行检测方法

冀立光, 周蓓, 杨鸿儒, 周玉畅, 崔梦琦, 许瑾晨   

  1. 信息工程大学网络空间安全学院 郑州 450001
  • 收稿日期:2025-02-05 修回日期:2025-04-26 发布日期:2026-02-10
  • 通讯作者: 许瑾晨(atao728208@126.com)
  • 作者简介:(1194868658@qq.com)

Parallel Detection Method of Maximum Floating-point Error Based on Gridding Particle SwarmOptimization Algorithm

JI Liguang, ZHOU Bei, YANG Hongru, ZHOU Yuchang, CUI Mengqi, XU Jinchen   

  1. School of Cyberspace Security,University of Information Engineering,Zhengzhou 450001,China
  • Received:2025-02-05 Revised:2025-04-26 Online:2026-02-10
  • About author:JI Liguang,born in 1992,postgraduate.His main research interest is high-performance computing.
    XU Jinchen,born in 1987,Ph.D,asso-ciate professor.His main research in-terest is high-performance computing.

摘要: 浮点计算程序广泛应用于航空航天、人工智能、国防军事、金融结算等领域,浮点程序的计算精度和性能直接关系到相关应用的安全和效果。最大浮点误差值是衡量浮点计算程序精度的核心关键指标,浮点误差的累积效应也会导致难以承受的灾难,因此需要研发一款精准高效的浮点数最大误差检测工具,为研究人员及时采取优化和干预措施提供支撑作用。对此,将浮点数最大误差检测问题转换为目标函数最大值搜索问题,充分发挥国产申威平台的主从架构两级并行计算模式的算力优势,深度挖掘粒子群启发式搜索算法的性能和精度潜能,采用“网格搜索、独立培养、分层汇聚、动态适应”的思想优化粒子群算法,根据搜索过程所处的不同阶段针对性地设置相关搜索参数,使得改进后的算法在搜索精度和搜索性能两个方面均有所提高。该算法为精确检测浮点数最大误差提供了一种新的实用工具和思路参考,同时进一步丰富了国产申威平台的工具库。

关键词: 浮点数, 误差检测, 粒子群优化算法, 并行计算, 申威平台

Abstract: Floating-point computing programs are widely used in aerospace,artificial intelligence,national defense and military,financial settlement and other fields.The computing accuracy and performance of floating-point programs are directly related to the safety and effectiveness of related applications.The maximum floating-point error is the key indicator to measure the accuracy of floating-point computing programs,and the cumulative effect of floating-point errors will also lead to unbearable disasters,so it is necessary to develop an accurate and efficient float-point maximum error detection tool to provide support for researchers to take timely optimization and intervention measures.The proposed algorithm transforms the problem of maximum error detection into the problem of searching for the maximum value of the objective function,gives full play to the computing power advantages of the master-slave architecture two-level parallel computing mode of the domestic Sunway platform,deeply excavates the performance and accuracy potential of the particle swarm heuristic search algorithm,and optimizes the particle swarm algorithm with the idea of grid search,independent cultivation,hierarchical convergence and dynamic adaptation.According to the different stages of the search process,the relevant search parameters are set,so that the improved algorithm achieves improvement in both search accuracy and search performance.This provides a new practical tool and thinking reference for accurately detecting the maximum error of floating-point numbers,and further enriches the tool library of domestic Sunway platform.

Key words: Floating-point, Error detection, Particle swarm optimization algorithm, Parallel computing, Sunway platform

中图分类号: 

  • TP314
[1]MACHIANI H N,TALEIZADEH A A,TOLOO M,et al.Designing a new sustainable healthcare network considering the COVID-19 pandemic:Artificial intelligence-based solutions[J].Expert Systems with Applications,2025,260:125357-125357.
[2]NING D.Computer Software Design Based on Cloud Platform High-Performance Computing[J].Journal of Physics:Confe-rence Series,2021,1915(3):032005.
[3]LUСKIJ G,DOLHOLENKO O.Development of floating pointoperating devices[J].Technology Audit and Production Reserves,2023,5(2):11-17.
[4]OHTA Y,OZAKI K.Extension of floating-point filters to absolute and relative errors for numerical computation[J].Journal of Physics:Conference Series,2019,1218(1):012011.
[5]UBLAIR M,OBENSKI S,BRIDICKAS P.Patriot missile de-fense:software problem led to system failure at Dhahran,Saudi Arabia[R].Washington:United States Government Accountability Office,1992.
[6]Wikipedia.Ariane-5 flight 50 1[EB/OL].http://en.wikipedia.Org/wiki/Ariane-5-Flight_501.
[7]CNN.Toyota:Software to blame for Prius brake problems[EB/OL].http://edition.cnn.com/2010/WORLD/asiapcf/02/04/japan.prius.complaints/.
[8]BAGNARA R,BAGNARA A,BISELLI F,et al.Correct ap-proximation of IEEE 754 floating-point arithmetic for program verification[J].Constraints,2022,27(1/2):29-69.
[9]YI X,CHEN L,MAO X,et al.Efficient automated repair of high floating-point errors in numerical libraries[J].Proceedings of the ACM on Programming Languages,2019,3:1-29.
[10]SARMA R,BHARGAVA C,KOTECHA K,et al.An Evolu-tionary Normalization Algorithm for Signed Floating-Point Multiply-Accumulate Operation[J].Computers,Materials & Continua,2022,72(1):481-495.
[11]ZOU D M.Search-oriented error testing and analysis of floating-point programs[D].Beijing:Peking University,2020.
[12]CHIANG W F,GOPALAKRISHNAN G,RAKAMARIC Z,et al.Efficient search for inputs causing high floating-point errors[C]//ACM SIGPLAN Notices.ACM,2014:43-52.
[13]CHILENSKIJ J,MILLER S P.Applicability of modified condition/decision coverage to software testing[J].Software Engineering Journal,1994,9(5):193-200.
[14]ZOU D,WANG R,XIONG Y,et al.A genetic algorithm for detecting significant floating- point inaccuracies[C]//Proceedings of the 37th International Conference on Software Engineering.IEEE,2015:529-539.
[15]FU Z,BAI Z,SU Z.Automated backward error analysis for numerical code[C]//Proceedings of the 2015 ACM SIGPLAN International Conference on Object-Oriented Programming,Systems,Languages,and Applications.ACM,2015:639-654.
[16]YIN X,CHEN L,MAO X,et al.Efficient Automated Repair of High Floating-Point Errors in Numerical Libraries[C]//Proceedings of the ACM on Programming Languages.2019.
[17]BARR E T,VO T,LE V,et al.Automatic detection of floating-point exceptions[C]//ACM SIGPLAN Notices.2013:549-560.
[18]HUAYU F,DIAN L,HAIBING H,et al.A fast PSO algorithm based on Alpha-stable mutation and its application in aerodynamic optimization[J].Xibei Gongye Daxue Xuebao/Journal of Northwestern Polytechnical University,2022,40(6):1385-1393.
[19]MARCELA B S S,ARRIGO C,ALBERTOC T.Genetic algo-rithm with a Bayesian approach for multiple change-point detection in time series of counting exceedances for specific thre-sholds[J].Journal of the Korean Statistical Society,2023,52(4):982-1024.
[20]SEIFOLLAHI S,BAGIROV A,BORZESHI Z E,et al.A simulated annealing-based maximum-margin clustering algorithm[J].Computational Intelligence,2019,35(1):23-41.
[21]EBERHART R C,KENNEDY J.A New Optimizer Using Particle Swarm Theory[C]//Proceedings of the Sixth International Symposium on Micro Machine and Human Science.Nagoya,1995:39-43.
[22]KENNEDY J,EBERHART R C.Particle Swarm Optimization[C]//Proceeding of IEEE International Conference on Neural Networks.1995:1942-1948.
[23]MULLER J M,BRUNIE N,DE DINECHIN F,et al.Handbook of Floating-Point Arithmetic[M].Springer International Publishing,2018.
[24]SOLOVYEV A,JACOBSEN C,RAKAMARIC′ Z,et al.Rigo-rous Estimation of Floating-Point Round-off Errors with Symbolic Taylor Expansions[C]//In 20th International Symposium on Formal Methods.New York:ACM,2015:532-550.
[25]CHIANG W F,BARANOWSKI M,BRIGGS I,et al.Rigorous floating-point mixed-precision tuning[C]//Symposium on Principles of Programming Languages.New York:ACM,2017:300-315.
[26]LIU L X.High performance data processing of distributed database and multi-core processor based on particle swarm optimization[J].Journal of Electronics and Information Science,2023,8(4):45-51.
[27]SUGANTHAN P N.Particle Swarm Optimizer with Neighborhood Operator[C]//Proceedings of Congress on Evolutionary Computation.1999:1958-1962.
[28]PANCHEKHA P,SANCHEZ-STERN A,WILCOX J R,et al.Automatically improving accuracy for floating point expressions[J].ACM SIGPLAN Notices,2015,50(6):1-11.
[29]ZHANG Z Y,XU J C,HAO J W,et al.Hierarchical search algorithm for error detection in floating-point arithmetic expressions[J].The Journal of Supercomputing,2023,80:1183-1205.
[30]CATTANEO D,BELLO A D,CHERUBIN S,et al.Embedded Operating System Optimization through Floating to Fixed Point Compiler Transformation[C]//21st Euromicro Conference on Digital System Design.Institute of Electrical and Electronics Engineers,2018:172-176.
Viewed
Full text


Abstract

Cited

  Shared   
  Discussed   
No Suggested Reading articles found!