DocumentCode :
3580570
Title :
A Study of Task Scheduling Based on Differential Evolution Algorithm in Cloud Computing
Author :
Jing Xue ; Liutao Li ; SaiSai Zhao ; Litao Jiao
Author_Institution :
Sch. of Autom., Northwestern Polytech. Univ., Xi´an, China
fYear :
2014
Firstpage :
637
Lastpage :
640
Abstract :
In this paper, we put forward a task scheduling algorithm in cloud computing with the goal of the minimum completion time, maximum load balancing degree, and the minimum energy consumption using improved differential evolution algorithm. In order to improve the global search ability in the earlier stage and the local search ability in the later stage, we have adopted the adaptive zooming factor mutation strategy and adaptive crossover factor increasing strategy. At the same time, we have strengthened the selection mechanism to keep the diversity of population in the later stage. In the process of simulation, we have performed the functional verification of the algorithm and compared with the other representative algorithms. The experimental results show that the improved differential evolution algorithm can optimize cloud computing task scheduling problems in task completion time, load balancing, and energy efficient optimization.
Keywords :
cloud computing; evolutionary computation; scheduling; task analysis; adaptive crossover factor; adaptive zooming factor mutation strategy; cloud computing task scheduling problems; differential evolution algorithm; energy efficient optimization; functional verification; global search ability; local search ability; maximum load balancing degree; minimum completion time; minimum energy consumption; representative algorithms; selection mechanism; task completion time; task scheduling algorithm; Cloud computing; Energy consumption; Scheduling; Scheduling algorithms; Sociology; Statistics; Cloud Computing; Differential Evolution Algorithm; Task Scheduling;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Computational Intelligence and Communication Networks (CICN), 2014 International Conference on
Print_ISBN :
978-1-4799-6928-9
Type :
conf
DOI :
10.1109/CICN.2014.142
Filename :
7065562
Link To Document :
بازگشت