Towards Order-Preserving and Zero-Copy Communication on Shared Memory for Large Scale Simulation

被引:1
作者
Li Xiuhe [1 ]
Shen Yang [1 ]
Lin Zhongwei [1 ]
Zhao Shunkai [1 ]
Shi Qianqian [1 ]
Dai Shaoqi [1 ]
机构
[1] Natl Univ Def Technol, Coll Elect Engn, Hefei 230031, Peoples R China
基金
中国国家自然科学基金;
关键词
Zero-copy; Order-preserving; Shared memory; Parallel simulation; Message passing interface; Interprocess communicaton;
D O I
10.23919/cje.2021.00.393
中图分类号
TM [电工技术]; TN [电子技术、通信技术];
学科分类号
0808 ; 0809 ;
摘要
Parallel simulation generally needs efficient, reliable and order-preserving communication. In this article, a zero-copy, reliable and order-preserving intra-node message passing approach named ZeROshm is proposed. This mechanism partitions shared memory into segments assigned to processes for receiving messages. Each segment consists of two levels of index L1 and L2 that recordes the order of messages in the host segment, and the processes also read from and write to the segments directly according to the indexes, thereby eliminating allocating and copying buffers. As experimental results show, ZeROshm exhibits nearly equivalent performance to message passing interface (MPI) for small message and superior performance for large message. Specifically, ZeROshm costs less time by 43%, 40% and 55% respectively in pure communication, communication with contention and real PHOLD simulation within a single node. Additionally, in hybrid environment, the combination of ZeROshm and MPI also shorten the execution time of PHOLD simulation by about 42% compared to pure MPI.
引用
收藏
页码:1066 / 1076
页数:11
相关论文
共 18 条
[1]   Shared memory consistency models: A tutorial [J].
Adve, SV ;
Gharachorloo, K .
COMPUTER, 1996, 29 (12) :66-&
[2]  
[Anonymous], 2015, P 3 ACM SIGSIM C PRI
[3]  
Aublin Pierre-Louis, 2011, Technical Report
[4]  
Forum M., 2021, MPI: A Message-Passing Interface Standard Version 4.0
[5]  
Fujimoto R.M, 1999, Parallel and Distribution Simulation Systems, V1st
[6]   Research Challenges in Parallel and Distributed Simulation [J].
Fujimoto, Richard M. .
ACM TRANSACTIONS ON MODELING AND COMPUTER SIMULATION, 2016, 26 (04)
[7]  
Hoeer Torsten., 2012, Proceedings of the 19th European Conference on Recent Advances in the Message Passing Interface, EuroMPI'12, P132, DOI DOI 10.1007/978-3-642-33518-1_18
[8]  
Hoefler T, 2010, LECT NOTES COMPUT SC, V6305, P132, DOI 10.1007/978-3-642-15646-5_14
[9]   Optimization of Parallel Discrete Event Simulator for Multi-core Systems [J].
Jagtap, Deepak ;
Abu-Ghazaleh, Nael ;
Ponomarev, Dmitry .
2012 IEEE 26TH INTERNATIONAL PARALLEL AND DISTRIBUTED PROCESSING SYMPOSIUM (IPDPS), 2012, :520-531
[10]  
KRANZ D, 1993, SIGPLAN NOTICES, V28, P54, DOI 10.1145/173284.155338