B. A. Malloy and A. T. Montroy. A parallel distributed simulation of a large-scale PCS network: Keeping secrets. In C. Alexopoulos, K. Kang, W. R. Lilegdon, and D. Goldsman, editors, Proceedings of the 1995.  Home/Search   Document Not in Database   Summary   Related Articles   Check

....protocols have a wider application range than lookahead based protocols do. In this paper, we classify lookback into four types, and show that all four exist in the PCS (Personal Communication Service) network simulation. PCS network simulation has been widely studied by the PDES community [1, 8, 11, 12], yet it is difficult to simulate it efficiently using lookahead based conservative protocols. In this paper, we demonstrate that lookback based protocol is capable of efficient parallel PCS network simulations. Moreover, we demonstrate that different types of lookback can be exploited to ....

B. A. Malloy and A. T. Montroy. A parallel distributed simulation of a large-scale PCS network: Keeping secrets. In C. Alexopoulos, K. Kang, W. R. Lilegdon, and D. Goldsman, editors, Proceedings of the 1995.

....show that all four types of lookback exist in the PCS network simulation, exploitable by either lookback based protocols or optimistic protocols. 8. 1 Exploiting Lookback in PCS Network Simulation A classical subject of the PDES research, the PCS network simulation has attracted many researchers [14, 11, 45, 32], because it actually represents a class of more general systems where various mobile objects roaming around a geographically divided domain compete for limited amount of resources in each location. Many real systems can be abstracted in this way, so the study of PCS simulation may have ....

Brian A. Malloy and Albert T. Montroy. A parallel distributed simulation of a large-scale PCS network: Keeping secrets. In C. Alexopoulos, K. Kang, W. R. Lilegdon, and D. Goldsman, editors, Proceedings of the 1995.

....presented in this paper concentrates on the simulation of a class of problems known as spatially explicit,problems that have a structure which naturally contains spatial information. Among such problems are battlefield simulations [35, 18] the simulation of Personal Communications Services (PCS) [7, 16, 23, 3] etc. In general, spatially explicit applications consist of a multidimensional space, which is discretized into a multidimensional lattice. Each lattice node can contain information which characterizes a given area. There Parts of this work were published in the Proceedings of Parallel and ....

....only the Move Event is non local. Because of the discrete nature of individual based models such as the one described above, they lend themselves well to discrete event simulation techniques. Discrete Event Simulation (DES) is a technique which has been used in modeling communication systems[7, 16, 23, 3], electronic circuits [10] battlefield scenarios [35, 18] etc. In DES the physical system is modeled by a single process consisting of the process state, the clock, and the event queue. Events are kept in a priority queue, where the event with the lower scheduled time has higher priority. The ....

B. A. Malloy and A. T. Montroy. A Parallel Distributed Simulation of a Large-Scale PCS Network: Keeping Secrets. Winter Simulation Conference, pages 571-578, 1995.

....and so the small lookahead which is adequate for GloMoSim s detailed radio simulations is inadequate for us. GloMoSim and SWAN are among the most current re ections of work on parallelized wireless simulation. Interest in the area has been active in recent years, examples of relevant work are [8, 9, 10, 12, 20, 22, 15, 19]. Emphasis and problems considered vary from project to project. A common theme in the list just cited is an interest in simulation of cellular communications. This problem has di erent radio characteristics than does ad hoc sensor networks, principally because sensor networks use very low power ....

B. A. Malloy and A. T. Montroy. A parallel distributed simulation of a large-scale PCS network: keeping secrets. Proceedings of

....18, 19, 20] the e ect of di erent call models and portable movement patterns are studied. The models are implemented using distributed optimistic Time Warp simulation mechanism. There are similar studies that make use of either optimistic approach [116] or conservative synchronization approach [74, 11, 12, 13]. Albeit important in studies of general network performance and capacity, the above teletrac models assumes a great simpli cation of the real systems in which radio channel models that include signal propagation and interference e ects are not studied in detail. Modeling increasing level of ....

B. A. Malloy and A. T. Montroy. A parallel distributed simulation of a large-scale PCS network: keeping secrets. Proceedings of

....the broadcast area of a single base station. A set of radio frequencies is allocated to the cell that can be used by mobiles to send and receive calls while in the cell. In cell based simulation, the cell is a predefined geographic region usually represented by a square or hexagon (e.g. 1] [5]) As mobile units cross from one cell to another a process called hand off occurs. The hand off transfers the mobile unit from one base station to another. By incorporating a signal propagation model (based on [4] the shape of the cell is defined by the signal strength equation instead of being ....

B. Malloy and A. Montroy. A Parallel Distributed Simulation of a Large-Scale PCS Network: Keeping Secrets. In Proceedings of the 1995 Winter Simulation Conference, Arlington, Virginia, 1995.

....technology is advancing rapidly. Thus, the simulation model must be adaptable to allow advances in technology to be easily incorporated. In previous work, we have addressed the first problem by parallelizing the PCS simulation using a conservative protocol on a distributed network of workstations[8, 9]. By using a unique look ahead technique called secrets and by distributing the computation and memory references across eight workstations, our speedups ranged from 3.4 to 7.5 over the fastest sequential execution, reducing the time to perform the simulation from over 20 hours to less than 4 ....

Albert M. Montroy and Brian A. Malloy. A parallel distributed simulation of a large-scale PCS network: Keeping secrets. Technical Report 95-106, Clemson University, May 1995.

....calls located in the cell may use those channels. In dynamic channel allocation, the channel being used by a call may actually be located in a nearby cell. of the network (boundary effect) 7] It has been shown that a network must contain more than a thousand cells to avoid the boundary effect[1, 8]. A second problem is that PCS technology is advancing rapidly. Thus, the simulation model must be adaptable to allow advances in technology to be easily incorporated. In previous work, we have addressed the first problem by parallelizing the PCS simulation using a conservative protocol on a ....

....technology is advancing rapidly. Thus, the simulation model must be adaptable to allow advances in technology to be easily incorporated. In previous work, we have addressed the first problem by parallelizing the PCS simulation using a conservative protocol on a distributed network of workstations[8, 9]. By using a unique look ahead technique called secrets and by distributing the computation and memory references across eight workstations, our speedups ranged from 3.4 to 7.5 over the fastest sequential execution, reducing the time to perform the simulation from over 20 hours to less than 4 ....

[Article contains additional citation context not shown here]

B. A. Malloy and A. M. Montroy. A parallel distributed simulation of a large-scale PCS network: Keeping secrets. Proceedings of the 1995 Winter Simulation Conference, December 1995. To appear.

No context found.

B. Malloy and A. Montroy, A parallel distributed simulation of a large-scale PCS network: keeping secrets, in: Proceedings of

No context found.

Malloy B. and Montroy A.: "A parallel distributed simulation of a large-scale PCS network: keeping secrets", Proceedings of the 1995 Winter Simulation Conference, pp. 571-578.

Online articles have much greater impact   More about CiteSeer.IST   Add search form to your site   Submit documents   Feedback  

CiteSeer.IST - Copyright Penn State and NEC