Towards an Architecture Proposal for Federation of Distributed DES Simulators

The simulation of large and complex Discrete Event Systems (DESs) increasingly imposes more demanding and urgent requirements on two aspects accepted as critical: (1) Intensive use of models of the simulated system that can be exploited in all phases of its life cycle where simulation can be used, and methodologies for these purposes; (2) Adaptation of simulation techniques to HPC infrastructures, as a method to improve simulation efficiency and to have scalable simulation environments. This paper proposes a Model Driven Engineering approach (MDE) based on Petri Nets (PNs) as formal model. This approach proposes a domain specific language based on modular PNs from which efficient distributed simulation code is generated in an automatic way. The distributed simulator is constructed over generic simulation engines of PNs, each one containing a data structure representing a piece of net and its simulation state. The simulation engine is called simbot and versions of it are available for different platforms. The proposed architecture allows, in an efficient way, a dynamic load balancing of the simulation work because the moving of PN pieces can be realized by moving a small number of integers representing the subnet and its state.

[1]  Patrizio Dazzi,et al.  A Holistic Approach for High-level Programming of Next-generation Data-intensive Applications Targeting Distributed Heterogeneous Computing Environment , 2016, Cloud Forward.

[2]  Mohamed Bettaz,et al.  Performance comparison of high-level algebraic nets distributed simulation protocols , 1998, J. Syst. Archit..

[3]  Dirk Muthig,et al.  A Scalable, Reactive Architecture for Cloud Applications , 2018, IEEE Software.

[4]  David M. Nicol,et al.  Automated Parallelization of Timed Petri-Net Simulations , 1993, J. Parallel Distributed Comput..

[5]  Netsanet Haile,et al.  Evaluating investments in portability and interoperability between software service platforms , 2018, Future Gener. Comput. Syst..

[6]  Vlado Stankovski,et al.  QoS-Aware Orchestration of Network Intensive Software Utilities within Software Defined Data Centres , 2018, Journal of Grid Computing.

[7]  Kalyan S. Perumalla,et al.  /spl mu/sik - a micro-kernel for parallel/distributed simulation systems , 2005, Workshop on Principles of Advanced and Distributed Simulation (PADS'05).

[8]  Philipp Haller,et al.  On the integration of the actor model in mainstream technologies: the scala perspective , 2012, AGERE! 2012.

[9]  Gul Agha,et al.  Concurrent Object-Oriented Programming and Petri Nets , 2001, Lecture Notes in Computer Science.

[10]  Azzedine Boukerche,et al.  Optimized Federate Migration for Large-Scale HLA-Based Simulations , 2008, 2008 12th IEEE/ACM International Symposium on Distributed Simulation and Real-Time Applications.

[11]  José A. Bañares,et al.  Model and Simulation Engines for Distributed Simulation of Discrete Event Systems , 2018, GECON.

[12]  Bernard P. Zeigler,et al.  Theory of Modeling and Simulation: Integrating Discrete Event and Continuous Complex Dynamic Systems , 2000 .

[13]  Rafael Tolosana-Calasanz,et al.  Model-driven development of data intensive applications over cloud resources , 2018, Future Gener. Comput. Syst..

[14]  Sai Peck Lee,et al.  A semantic interoperability framework for software as a service systems in cloud computing environments / Reza Rezaei , 2014 .

[15]  José Merseguer,et al.  Towards a UML profile for data intensive applications , 2016, QUDOS@ISSTA.

[16]  José Manuel Colom,et al.  Implementation of Weighted Place/Transition Nets Based on Linear Enabling Functions , 1994, Application and Theory of Petri Nets.

[17]  Azzedine Boukerche,et al.  Dynamic balancing of communication and computation load for HLA-based simulations on large-scale distributed systems , 2011, J. Parallel Distributed Comput..

[18]  David M. Nicol,et al.  Parallel discrete event simulation: The making of a field , 2017, 2017 Winter Simulation Conference (WSC).

[19]  Richard M. Fujimoto,et al.  Research Challenges in Parallel and Distributed Simulation , 2016, ACM Trans. Model. Comput. Simul..