DocumentCode :
2936402
Title :
Parallel simulation of the Sharks World problem
Author :
Bagrodia, Rajive L. ; Liao, Wen-Toh
Author_Institution :
Dept. of Comput. Sci., California Univ., Los Angeles, CA, USA
fYear :
1990
fDate :
9-12 Dec 1990
Firstpage :
191
Lastpage :
198
Abstract :
The authors describe the design of a Maisie program to simulate the Shark´s World problem and present the timings for both sequential and parallel executions of the model. The sequential implementation of Maisie used a single event-list and a global simulation clock. The parallel implementation was executed on a multicomputer architecture using a space-time simulation algorithm. Experimental results indicate that reasonable speedups may be achieved for a relatively small number of creatures; further, the speedups increased linearly as the number of creatures in the system was increased. For a configuration with 2048 creatures, the parallel implementation with 32 nodes and 32 sectors was better by a factor of almost five over the best sequential implementation. The performance of the space-time algorithm depends on a number of factors, including frequency of checkpointing, time-length for which multiple incarnations of an entity are created, and the algorithm used to detect the time up to which the simulation has converged
Keywords :
discrete event simulation; parallel processing; Maisie program; Sharks World; checkpointing; global simulation clock; multicomputer architecture; multiple incarnations; sequential implementation; single event-list; space-time algorithm; space-time simulation algorithm; time-length; Algorithm design and analysis; Application software; Clocks; Computational modeling; Computer science; Concurrent computing; Discrete event simulation; Memory architecture; Parallel architectures; Resumes;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Simulation Conference, 1990. Proceedings., Winter
Conference_Location :
New Orleans, LA
Print_ISBN :
0-911801-72-3
Type :
conf
DOI :
10.1109/WSC.1990.129513
Filename :
129513
Link To Document :
بازگشت