MPI/OpenMP-Based Parallel Solver for Imprint Forming Simulation

作     者：Yang Li Jiangping Xu Yun Liu Wen Zhong Fei Wang 

作者机构：School of Mechanical EngineeringJiangsu UniversityZhenjiang212016China School of Mechanical EngineeringWuhan Polytechnic UniversityWuhan430023China Shenyang Mint Company LimitedShenyang110092China 

出 版 物：《Computer Modeling in Engineering & Sciences》 (工程与科学中的计算机建模（英文）)

年 卷 期：2024年第140卷第7期

页      面：461-483页

核心收录：

学科分类：08[工学] 0835[工学-软件工程] 0701[理学-数学] 081202[工学-计算机软件与理论] 0812[工学-计算机科学与技术（可授工学、理学学位）] 

基　　金：supported by the fund from ShenyangMint Company Limited(No.20220056) Senior Talent Foundation of Jiangsu University(No.19JDG022) Taizhou City Double Innovation and Entrepreneurship Talent Program(No.Taizhou Human Resources OfficeNo.22) 

主　　题：Hybrid MPI/OpenMP parallel computing MPI OpenMP imprint forming 

摘      要：In this research,we present the pure open multi-processing(OpenMP),pure message passing interface(MPI),and hybrid MPI/OpenMP parallel solvers within the dynamic explicit central difference algorithm for the coining process to address the challenge of capturing fine relief features of approximately 50 *** such precision demands the utilization of at least 7 million tetrahedron elements,surpassing the capabilities of traditional serial programs previously *** mitigate data races when calculating internal forces,intermediate arrays are introduced within the OpenMP *** helps ensure proper synchronization and avoid conflicts during parallel ***,in the MPI implementation,the coins are partitioned into the desired number of *** division allows for efficient distribution of computational tasks across multiple *** simulation examples are conducted to compare the three solvers with serial programs,evaluating correctness,acceleration ratio,and parallel *** results reveal a relative error of approximately 0.3%in forming force among the parallel and serial solvers,while the predicted insufficient material zones align with experimental ***,speedup ratio and parallel efficiency are assessed for the coining process *** pureMPI parallel solver achieves a maximum acceleration of 9.5 on a single computer(utilizing 12 cores)and the hybrid solver exhibits a speedup ratio of 136 in a cluster(using 6 compute nodes and 12 cores per compute node),showing the strong scalability of the hybrid MPI/OpenMP programming *** approach effectively meets the simulation requirements for commemorative coins with intricate relief patterns.