Improving performance by replicating simulations with alternative synchronization approaches

Parallel and distributed simulation facilitates the construction of a simulation application (i.e., federation in HLA terminology) with a number of simulation components (federates). Recently, an approach based on active replication technique has been proposed to improve the performance of simulations by exploring software diversity. To guarantee the correctness of the approach, all replicas of the same federate are required to be Piece-Wise-Deterministic (PWD). However, the PWD restriction will not be satisfied if the replicas achieve software diversity by employing different kinds of synchronization approaches. A replication structure is proposed in this paper, which can transparently handle the federate replicas that use either conservative or optimistic synchronization approach. The execution speed of the federation is increased by always choosing the fastest replica to represent the federate in the federation. Besides presenting the implementation details, we also report the experimental results to demonstrate the performance improvement of the replication structure.

[1]  Hassan Rajaei,et al.  The local Time Warp approach to parallel simulation , 1993, PADS '93.

[2]  Ian T. Foster,et al.  The anatomy of the grid: enabling scalable virtual organizations , 2001, Proceedings First IEEE/ACM International Symposium on Cluster Computing and the Grid.

[3]  Ewa Niewiadomska-Szynkiewicz,et al.  Algorithms for distributed simulation - comparative study , 2002, Proceedings. International Conference on Parallel Computing in Electrical Engineering.

[4]  Leon F. McGinnis,et al.  Optimistic-Conservative Synchronization in Distributed Factory Simulation , 2006, Proceedings of the 2006 Winter Simulation Conference.

[5]  K. Mani Chandy,et al.  Distributed Simulation: A Case Study in Design and Verification of Distributed Programs , 1979, IEEE Transactions on Software Engineering.

[6]  JhaVikas,et al.  A unified framework for conservative and optimistic distributed simulation , 1994 .

[7]  Stephen John Turner,et al.  A Service Oriented HLA RTI on the Grid , 2007, IEEE International Conference on Web Services (ICWS 2007).

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

[9]  Richard M. Fujimoto,et al.  Parallel and Distribution Simulation Systems , 1999 .

[10]  Francesco Quaglia.  A Middleware Level Active Replication Manager for High Performance HLA-based Simulations on SMP Systems , 2006, 2006 Tenth IEEE International Symposium on Distributed Simulation and Real-Time Applications.

[11]  Behrokh Samadi.  Distributed simulation, algorithms and performance analysis (load balancing, distributed processing) , 1985 .

[12]  R.M. Fujimoto,et al.  Parallel and distributed simulation systems , 2001, Proceeding of the 2001 Winter Simulation Conference (Cat. No.01CH37304).

[13]  ThorelliLars-Erik,et al.  The local Time Warp approach to parallel simulation , 1993 .

[14]  Stephen John Turner,et al.  A Hybrid HLA Time Management Algorithm Based on Both Conditional and Unconditional Information , 2008, 2008 22nd Workshop on Principles of Advanced and Distributed Simulation.

[15]  Randal E. Bryant,et al.  SIMULATION OF PACKET COMMUNICATION ARCHITECTURE COMPUTER SYSTEMS , 1977 .

[16]  Rajive L. Bagrodia,et al.  A unified framework for conservative and optimistic distributed simulation , 1994, PADS '94.