P-HASE: An Efficient Synchronous PDES Tool for Creating Scalable Simulations
Synchronous, parallel discrete event simulation (PDES) is the simplest and lightweight approach to speedup large-scale simulations by scheduling as many events, of the same simulation cycle, to be executed concurrently. The scheduling technique to achieve perfect load balance and scalability is a key challenge for an efficient synchronous PDES. In this paper, we proposed a technique for balancing loads to fit the number of available processors on multicores. The technique has been implemented on a synchronous PDES tool called P-HASE (the Parallel - Hierarchical computer Architecture design and Simulation Environment) using the NET 4.0 concurrency runtime and OpenMP. Eight simulation models have been evaluated on 4-, 8-, and 16- core machines. The results show that the models using P-HASEare faster than HASE for 18 – 6.5 times; and maintain their performance when changing the numbers of processors. The results confirm that the simulation models created by using the P-HASE tool are highly scalable for multicore architecture.
KeywordsSynchronous Parallel Discrete Event Simulation PDES Framework Scalable Modelling Task-based Parallelism .NET
Unable to display preview. Download preview PDF.
- 1.Fujimoto, R.M.: Parallel discrete event simulation. In: Proceedings of the 21st Conference on Winter Simulation, pp. 19–28. ACM, Washington, D.C. (1989)Google Scholar
- 2.Lim, C.-C., Gan, B.-P., Low, Y.-H.: Computing Safetime in a Conservative Synchro-nous Simulation Based on Future Events. In: PDPTA 1999, pp. 2436–2442 (1999)Google Scholar
- 6.Weingärtner, E., et al.: SliceTime: a platform for scalable and accurate network emulation. In: Proceedings of the 8th USENIX Conference on Networked Systems Design and Implementation, pp. 19–19. USENIX Association, Boston (2011)Google Scholar
- 7.Fujimoto, R.M.: Parallel and distributed simulation. In: Proceedings of the 31st Conference on Winter Simulation: Simulation—a Bridge to the Future, vol. 1, pp. 122–131. ACM, Phoenix (1999)Google Scholar
- 9.Over, A., Clarke, B., Strazdins, P.: A Comparison of Two Approaches to Parallel Simulation of Multiprocessors. In: IEEE International Symposium on Performance Analysis of Systems & Software, ISPASS 2007, pp. 12–22 (2007)Google Scholar
- 10.Park, H., Fishwick, P.A.: A fast hybrid time-synchronous/event approach to parallel discrete event simulation of queuing networks. In: Proceedings of the 40th Conference on Winter Simulation 2008, Miami, Florida, pp. 795–803 (2008)Google Scholar
- 12.Mackey, A.: Parallelization and Threading Enhancements. In: Introducing .NET 4.0, pp. 97–126. Apress (2010)Google Scholar
- 14.Sulistio, A., Yeo, C.S., Buyya, R.: A taxonomy of computer-based simulations and its mapping to parallel and distributed systems simulation tools. Software: Practice and Experience 34(7), 653–673 (2004)Google Scholar
- 15.Mallet, F., Alam, S., Ibbett, R.N.: An Extensible Clock Mechanism for Computer Architecture Simulations. In: Proceedings of the 13th IASTED International Conference on Modelling and Simulation 2002, Marina del Rey, California (2002)Google Scholar