Performance Improvement Using Parallel Simulation Protocol and Time Warp for DEVS Based Applications

DEVS is a formalism intended to model both discrete and continuous systems. The use of discrete events, rather than time steps, as the basis for simulation has been shown to reduce the computation time in many applications. Parallel DEVS is an extension to standard DEVS, which provides means to handle simultaneous scheduling. In this paper, we present an implementation of the parallel DEVS simulation protocol that uses a modified time warp optimistic algorithm for shared memory multiprocessor machine. This implementation is designed to execute the DEVS models in parallel and, at the same time to correctly simulate every model defined in terms of DEVS specification. Two test cases and the DEVSFIRE example are used to verify this algorithm. Preliminary experimental results are presented that show the implementation can speedup a DEVS simulation.

[1]  David R. Jefferson,et al.  Virtual time , 1985, ICPP.

[2]  Bernard P. Zeigler,et al.  Parallel DEVS: a parallel, hierarchical, modular modeling formalism , 1994, Proceedings of Winter Simulation Conference.

[3]  Fernando J. Barros,et al.  Modeling formalisms for dynamic structure systems , 1997, TOMC.

[4]  Bernard P. Zeigler,et al.  Distributed supply chain simulation in a DEVS/CORBA execution environment , 1999, WSC '99.

[5]  B. Zeigler,et al.  Support for Hierarchical Modular Component-based Model Construction in DEVS / HLA 1 , 1999 .

[6]  Ernesto Kofman,et al.  Quantized-state systems: a DEVS Approach for continuous system simulation , 2001 .

[7]  Gabriel A. Wainer,et al.  Application of the Cell-DEVS Paradigm for Cell Spaces Modelling and Simulation , 2001, Simul..

[8]  B. Zeigler,et al.  A High Performance Simulation Engine for Large-Scale Cellular DEVS Models , 2004 .

[9]  Bernard P. Zeigler,et al.  Forest Fire Spread and Suppression in DEVS , 2004, Simul..

[10]  Gabriel A. Wainer,et al.  DEVStone: a benchmarking technique for studying performance of DEVS modeling and simulation environments , 2005, Ninth IEEE International Symposium on Distributed Simulation and Real-Time Applications.

[11]  Norbert Giambiasi,et al.  Improving the lookahead computation in G-DEVS/HLA environment , 2005, Ninth IEEE International Symposium on Distributed Simulation and Real-Time Applications.

[12]  B. Zeigler,et al.  DEVS / RMI — AnAuto-Adaptive and Reconfigurable Distributed Simulation Environment for Engineering Studies , 2005 .

[13]  François E. Cellier,et al.  Quantised State System Simulation in Dymola/Modelica using the DEVS Formalism , 2006 .

[14]  Bernard P. Zeigler,et al.  Transforming DEVS to non-modular form for faster cellular space simulation , 2006 .

[15]  Xiaolin Hu,et al.  Performance Measurement of DEVS Dynamic Structure on Forest Fire Spreading Simulation , 2006 .

[16]  Barbara Chapman,et al.  Using OpenMP: Portable Shared Memory Parallel Programming (Scientific and Engineering Computation) , 2007 .

[17]  James J. Nutaro,et al.  On constructing optimistic simulation algorithms for the discrete event system specification , 2008, TOMC.