This paper presents the issues involved in the design and development of ussf. Parallel simulation techniques are used to enable optimal time versus resource tradeos in ussf. The techniques employed in the framework to reduce and regulate the memory requirements of the simulations are described. The API needed for model development is illustrated. The results obtained from the experiments conducted using various system models with two parallel simulation kernels (comparing a conventional approach with ussf) are also presented. Key Words: Large Scale Modeling, Parallel and Distributed Simulation, Time Warp Simulation, Unsynchronized Simulation 1 Support for this work was provided in part by the Defense Advanced Research Projects Agency under contract DABT63-96{C{0055. 1 2 RAO AND WILSEY 1. INTRODUCTION Modern systems such as microprocessors and communicat

Abstract not found

Time-Parallel Simulation offers the potential of massive parallelization of a simulation application, due to the amount of achievable parallelism not being restricted by the decomposability of the state space of a simulation model. Unfortunately, the potential speedup of a time-parallel simulation highly depends on the ability to match final and initial states of adjacent partitions. However, depending on the properties of the underlying simulation model, it might be feasible to accept a simulation iteration, even if the states of adjacent partitions do not match exactly. This leads to the concept of approximative state matching in time-parallel simulation, which is introduced in this paper. Experiments with a prototypical implementation of a simple simulation model show encouraging results in terms of simulation speedup and introduced error. 1

Abstract. The traditional technique to simulate physical systems modeled by partial differential equations is by means of a time-stepped methodology where the state of the system is updated at regular discrete time intervals. This method has inherent inefficiencies. Recently, we proposed [1] a new asynchronous formulation based on a discrete-event-driven (as opposed to time-driven) approach, where the state of the simulation is updated on a “need-to-be-done-only ” basis. Using a serial electrostatic implementation, we obtained more than two orders of magnitude speedup compared with traditional techniques. Here we examine issues related to the parallel extension of this technique and discuss several different parallel strategies. In particular, we present in some detail a newly developed discrete-event based parallel electromagnetic hybrid code and its performance using conservative synchronization on a cluster computer. These initial performance results are encouraging in that they demonstrate very good parallel speedup for large-scale simulation computations containing tens of thousands of cells, though overheads for inter-processor communication remain a challenge for smaller computations. 1

. Communication networks have steadily increased in size and complexity to meet the growing demands of applications. Simulations have been used to model and analyze modern communication networks. Modeling and simulation of networks involving thousands of nodes is hard due to their sheer size and complexity. Complete models of the ultralarge networks need to be simulated in order to conduct in-depth studies of scalability and performance. Parallel simulation techniques need to be eÆciently utilized to obtain optimal time versus resource tradeos. Due to the complexity of the system, it becomes critical that the design of such frameworks follow well established design principles such as object oriented (OO) design, so as to meet the diverse requirements of portability, maintainability, extensibility, and ease of use. This paper presents the issues involved in the design and implementation of an OO framework to enable parallel simulation of ultra-large communication networks. The OO techn...

One of the major overheads that prohibits the wide spread deployment of parallel discrete event simulation (PDES) is the need to synchronize the distributed processes in the simulation. Considerable investigations have been conducted to analyze and optimize the two widely used synchronization strategies, namely the conservative and the optimistic simulation paradigms. However, little attention has been focussed on the de-nition and strictness of causality. Does causality need to be preserved in all types of simulations? Previously, we had suggested an answer to this question. We had argued that signi cant performance gains can be achieved byreconsidering this de nition to decide if the parallel simulation really needs to subscribe to the preservation of causality. In this paper, we investigate this issue even more closely. An in depth analysis using several example simulation models is presented in this paper. In addition, a comparative analysis between unsynchronized and Time Warp simulation is presented. 1

On behalf of:

Discrete event simulations are widely used to study and analyze active and conventional networking architectures and protocols. Active networks must coexist and communicate with conventional networks to eectively utilize and extend the infrastructure of the Internet. Hence, large scale network simulations containing both conventional and active components should be conducted to study crucial scalability and performance issues. In this paper, a framework to enable parallel co-simulation of conventional and active networks is described. The framework integrates anse, a parallel Active Networks Simulation Environment, with NS , a popular sequential network simulator. Object oriented techniques that completely insulate the application modules from the modications have been employed for parallelizing NS in order to eliminate changes to the network models. This paper presents the design and implementation of the parallel co-simulation framework along with the results obtained from our cosi...

With the advent of standardization efforts such as the High Level Architecture (HLA) and Distributed Interactive Simulation (DIS), inter-operability of military simulation models has emerged as the chief design requirement. By enforcing strict conformance to the DIS and/or HLA, the Defense Modeling and Simulation Office (DMSO) has so far been able to "mix-and-match" different simulation models and frameworks to satisfy the military's simulation needs. However, by linking different legacy simulations (simulations previously designed to operate independently) together, the simulation now has to correctly handle multiple levels of detail in the interacting simulation entities. In addition, a given entity itself can be represented in different ways each with a different level of detail (also referred to as resolution or fidelity). A natural question to ask in this situation is why does the model require different resolutions? Why isn't one level of resolution sufficient to address all the ...

The model used in this report focuses on the analysis of ship waiting statistics and stock fluctuations under different arrival processes. However, the basic outline is the same: central to both models are a jetty and accompanying tankfarm facilities belonging to a new chemical plant in the Port of Rotterdam. Both the supply of raw materials and the export of finished products occur through ships loading and unloading at the jetty. Since disruptions in the plants production process are very expensive, buffer stock is needed to allow for variations in ship arrivals and overseas exports through large ships. Ports provide jetty facilities for ships to load and unload their cargo. Since ship delays are costly, terminal operators attempt to minimize their number and duration. Here, simulation has proved to be a very suitable tool. However, in port simulation models, the impact of the arrival process of ships on the model outcomes tends to be underestimated. This article considers three arrival processes: stock-controlled, equidistant per ship type, and Poisson. We assess how their deployment in a port simulation model, based on data from a real case study, affects the efficiency of the loading and unloading process. Poisson, which is the chosen arrival process in many client-oriented simulations, actually performs worst in terms of both ship delays and required storage capacity. Stock-controlled arrivals perform best with regard to ship delays and required storage capacity. In the case study two types of arrival processes were considered. The first type are the so-called stock-controlled arrivals, i.e., ship arrivals are scheduled in such a way, that a base stock level is maintained in the tanks. Given a base stock level of a raw material or ...