DocumentCode :
1886758
Title :
Dynamic load balancing with an improved with an improved spectral bisection algorithm
Author :
Van Driessche, Rafael ; Roose, Dirk
Author_Institution :
Dept. of Comput. Sci., Katholieke Univ., Leuven, Belgium
fYear :
1994
fDate :
23-25 May 1994
Firstpage :
494
Lastpage :
500
Abstract :
The efficient parallel execution of grid-oriented scientific calculations requires the partitioning of the grid in such a way that the work load is equally distributed over the processors of the parallel machine and that communication among the processors is minimized. For unstructured static grids, good partitions are obtained by applying the recursive spectral bisection heuristic to the interdependency graph of the grid. We describe how even in case of dynamically changing grids, grid-oriented problems can be formulated as graph partitioning problems for the purpose of load balancing. We further describe an alternative spectral bisection algorithm that yields better partitions than the standard algorithm for the graphs that model dynamic load balancing problems
Keywords :
graph theory; mesh generation; parallel algorithms; parallel machines; resource allocation; dynamic load balancing; dynamic load balancing problems; graph partitioning problems; grid partitioning; grid-oriented problems; grid-oriented scientific calculations; improved spectral bisection algorithm; interdependency graph; parallel execution; parallel machine; processor communication; recursive spectral bisection heuristic; spectral bisection algorithm; unstructured static grids; Computer science; Concurrent computing; Costs; Finite element methods; Grid computing; Iterative algorithms; Load management; Load modeling; Parallel machines; Partitioning algorithms;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Scalable High-Performance Computing Conference, 1994., Proceedings of the
Conference_Location :
Knoxville, TN
Print_ISBN :
0-8186-5680-8
Type :
conf
DOI :
10.1109/SHPCC.1994.296683
Filename :
296683
Link To Document :
بازگشت