B.L. Noble, Comput. & Commun. Res. Center, Washington Univ., St. Louis, MO, USA
G.D. Peterson, Comput. & Commun. Res. Center, Washington Univ., St. Louis, MO, USA
R.D. Chamberlain, Comput. & Commun. Res. Center, Washington Univ., St. Louis, MO, USA
Explores the performance of three synchronous discrete-event simulation algorithms: the global clock algorithm, the conservative lookahead algorithm, and speculative computation. We examine the effects of granularity and present empirical data to illustrate at what granularity the algorithm has reasonable performance. We also investigate two techniques for decreasing both synchronization and load imbalance. In addition, we examine how various execution platforms impact the performance of the simulation, providing empirical data from a network of workstations and a shared-memory multiprocessor. The impact of shared computational resources on simulation performance is also explored. The simulated system is a network of queues connected in a torus topology.
Index Terms:
discrete event simulation; synchronisation; software performance evaluation; parallel algorithms; queueing theory; resource allocation; shared memory systems; network topology; local area networks; synchronous parallel discrete-event simulation algorithms; performance; global clock algorithm; conservative lookahead algorithm; speculative computation; granularity; synchronization; load imbalance; execution platforms; workstation network; shared-memory multiprocessor; shared computational resources; queueing network; torus topology
Citation:
B.L. Noble, G.D. Peterson, R.D. Chamberlain, "Performance of synchronous parallel discrete-event simulation," hicss, pp.185, 28th Hawaii International Conference on System Sciences (HICSS'95), 1995
PDF
IEEE Xplore Subscribers
Digg
Furl
Spurl
Blink
Simpy
Google
Del.icio.us
Y!MyWeb