Title :
A Distributed Termination Detection Algorithm for Dynamic Asynchronous Systems
Author :
Johnson, Paul ; Mittal, Neeraj
Author_Institution :
Dept. of Comput. Sci., Univ. of Texas at Dallas, Richardson, TX, USA
Abstract :
Termination detection in distributed systems has been a popular problem of study. It involves determining whether a computation running on multiple nodes has ceased all its activities. A large number of termination detection algorithms have been proposed for static distributed systems in which the number of nodes present in the system is fixed and never changes during runtime. Recently, the termination detection problem has been investigated in the context of dynamic distributed systems in which individual nodes may join and/or leave the system at any time. In this paper, we propose an efficient algorithm for detecting termination of a computation in a dynamic, asynchronous, distributed system that allows nodes to join as well as leave the system while the computation is in progress. Our simulation results indicate that our algorithm has lower message complexity as well as lower detection latency than other comparable algorithms for solving the same problem.
Keywords :
distributed processing; detection latency; distributed termination detection algorithm; dynamic asynchronous systems; message complexity; static distributed systems; Change detection algorithms; Computational modeling; Computer science; Delay; Detection algorithms; Distributed computing; Grid computing; Peer to peer computing; Runtime; USA Councils; Asynchronous Systems; Dynamic Systems; Termination Detection; Tree-based Algorithm;
Conference_Titel :
Distributed Computing Systems, 2009. ICDCS '09. 29th IEEE International Conference on
Conference_Location :
Montreal, QC
Print_ISBN :
978-0-7695-3659-0
Electronic_ISBN :
1063-6927
DOI :
10.1109/ICDCS.2009.16