Title :
Grid resource scheduling based on improved differential evolution algorithms
Author :
Xue, Shengjun ; Li, Chan ; Yang, Ming ; Nie, Jing
Author_Institution :
Sch. of Comput. & Software, Nanjing Univ. of Inf. Sci. & Technol., Nanjing, China
Abstract :
The algorithms of grid resource scheduling is one of the key factor in promoting grid performance, and a smart algorithms would substantially promote the effective utilization of grid resources, as well as significantly reduce the implementation time. A lot of algorithms have been applied into resource scheduling, in which the evolutionary algorithms are more prominent. The shortages of traditional differential evolution algorithm were discussed in this paper, which was easily trapping in the local value and lead to premature convergence. In order to solve these problems, an improved differential evolution algorithm was proposed to control the scaling factor with the utilization of mutation operator. Experiment showed that, the improved differential evolution algorithm had a significant performance in the grid resource scheduling.
Keywords :
evolutionary computation; grid computing; scheduling; differential evolution algorithm; evolutionary algorithms; grid resource scheduling; mutation operator; premature convergence; scaling factor; smart algorithm; Biological cells; Evolution (biology); Evolutionary computation; Optimal scheduling; Processor scheduling; Scheduling; Software algorithms; Differential evolution; Grid resource; Mutation operator; Resource scheduling;
Conference_Titel :
Natural Computation (ICNC), 2010 Sixth International Conference on
Conference_Location :
Yantai, Shandong
Print_ISBN :
978-1-4244-5958-2
DOI :
10.1109/ICNC.2010.5584832