Title :
Adapting a partitioning-based heuristic load-balancing algorithm to heterogeneous computing environments
Author_Institution :
Dept. of Comput. Sci., Western Univ. of Timisoara, Romania
Abstract :
A graph partitioning-based heuristic load-balancing algorithm known as the largest-task-first-with-minimum-finish-time-and-available-communication-costs from EVAH package by M.J. Djomehri, et. al. (2003) is modified in order to be dynamically adapted to heterogeneous computing environments like a grid. An example is given to show the improvement.
Keywords :
distributed processing; graph theory; resource allocation; graph partitioning-based heuristic load-balancing; heterogeneous computing environment; largest-task-first-with-minimum-finish-time-and-available-communication-cost; Computational fluid dynamics; Costs; Grid computing; Heuristic algorithms; History; Load management; Mathematical programming; Packaging; Parallel processing; Partitioning algorithms;
Conference_Titel :
Symbolic and Numeric Algorithms for Scientific Computing, 2005. SYNASC 2005. Seventh International Symposium on
Print_ISBN :
0-7695-2453-2
DOI :
10.1109/SYNASC.2005.16
