Title :
Fault-tolerant distributed simulation
Author_Institution :
Dept. of Comput. Sci., Maryland Univ., College Park, MD, USA
fDate :
6/13/1905 12:00:00 AM
Abstract :
The author considers two kinds of problems that may appear in a distributed discrete-event simulation: the appearance of deadlocks due to unknown lookaheads and processor crashes. Both of the problems are solved by taking global snapshots of the distributed simulation efficiently. It is proved that the recorded global states of the distributed simulation are consistent and can thus be used by the deadlock breaking algorithm and for checkpointing.
Keywords :
"Fault tolerance","System recovery","Communication channels","Propagation delay","Computational modeling","Computer science","Educational institutions","Discrete event simulation","Computer crashes","Clustering algorithms"
Conference_Titel :
Simulation Conference, 1991. Proceedings., Winter
Print_ISBN :
0-7803-0181-1
DOI :
10.1109/WSC.1991.185668
