Title :
A task scheduling algorithm based on multi-core processors
Author :
Xiaozhong Geng ; Gaochao Xu ; Dan Wang ; Ying Shi
Author_Institution :
Dept. of Comput. Sci. & Technol., Jilin Univ., Changchun, China
Abstract :
With the development of multi-core technology, the number of processing core is growing dramatically. In order to make full use of these processing cores of such large number, this paper proposed a task scheduling algorithm on the basis of task duplication, which is composed of three steps of operations so that threads are allocated to processing cores more suitably. This algorithm not only increases the executive efficiency of task scheduling, but also can adjust scheduling sets according to the number of processing core. This algorithm diminishes communication overhead and keeps load balancing between cores, and meanwhile speedup ratio of parallel program is improved. The simulation experiment data shows that the algorithm can find near optimal solutions in reasonable time, and that it can find solutions in less time than genetic algorithms when the number of threads increases.
Keywords :
genetic algorithms; multiprocessing systems; parallel algorithms; parallel programming; processor scheduling; resource allocation; communication overhead; genetic algorithm; load balancing; multicore processor; multicore technology; near optimal solution; parallel program; processing core; task duplication; task scheduling algorithm; Genetic algorithms; Instruction sets; Multicore processing; Resource management; Scheduling; Scheduling algorithm; load balancing; multi-core processor; task duplication; task scheduling;
Conference_Titel :
Mechatronic Science, Electric Engineering and Computer (MEC), 2011 International Conference on
Conference_Location :
Jilin
Print_ISBN :
978-1-61284-719-1
DOI :
10.1109/MEC.2011.6025620