Automating parallel simulation using parallel time streams

Abstract

This paper describes a package for parallel steady-state stochastic simulation that was designed to overcome problems caused by long simulation times experienced in our ongoing research in performance evaluation of high-speed and integrated-services communication networks, while maintaining basic statistical rigors of proper analysis of simulation output data. The package, named AKAROA, accepts ordinary (nonparallel) simulation programs, and all further stages of stochastic simulation should be transparent for users. The package employs a new method of sequential estimation for the Multiple-Replications-in-Parallel scenario. All basic functions, including the transformation of originally nonparallel simulators into ones suitable for parallel execution, control of the precision of estimates, and stopping of parallel simulation processes when the required precision of the overall steady-state estimates is achieved, are automated. The package can be used on multiprocessor systems and/or heterogeneous computer networks, involving an arbitrary number of processors. The design issues, architecture, and implementation of AKAROA, as well as the results of its preliminary performance studies are presented.

Research Challenges in Parallel and Distributed Simulation

“…Large-scale simulations can have large-scale data requirements and produce massive amounts of output data, raising questions of I/O performance. Virtually all usage scenarios will call for multiple simulation runs (e.g., see Yau [1999]); techniques to improve the efficiency of multiple replicated runs, for example, by exploiting commonalities among these runs, are needed.…”

Web Services-Based Parallel Replicated Discrete Event Simulation for Large-Scale Simulation Optimization

“…Similarly, Perles et al [21] extend this approach onto a grid. To further reduce the computational load, Vakili [22] proposes the Single Clock Multiple Yau [23] uses parallel time streams. Our PRDES approach conducts parallel replications on the web.…”

Analysis of Dual-Bus Metropolitan Area Networks Using Distributed Quantitative Stochastic Simulation

“…Recognising the need for obtaining reliable results within a practical time, all numerical results were produced using AKAROA [26][27][28][29][30][31][32][33], an object-oriented simulation package developed by us for automated precision control of steady-state estimates and automated parallel execution of quantitative simulations. Slots in a DQDB network can be queue-arbitrated slots (allocated dynamically to stations using the distributed queuing mechanism [7,21]) or pre-arbitrated slots (reserved for specific connections by a bandwidth manager [8]); we assume that all slots are queue-arbitrated.…”