IJCATR Volume 6 Issue 4

Cost-Efficient Task Scheduling with Ant Colony Algorithm for Executing Large Programs In Cloud Computing

Fatemeh Imani , Shiva Razzaghzadeh , Masoud Bekravi
10.7753/IJCATR0604.1005
keywords : cloud computing; formatting; task scheduling; makespan; load balancing

PDF
The aim of cloud computing is to share a large number of resources and pieces of equipment to compute and store knowledge and information for great scientific sources. Therefore, the scheduling algorithm is regarded as one of the most important challenges and problems in the cloud. To solve the task scheduling problem in this study, the ant colony optimization (ACO) algorithm was adapted from social theories with a fair and accurate resource allocation approach based on machine performance and capacity. This study was intended to decrease the runtime and executive costs. It was also meant to optimize the use of machines and reduce their idle time. Finally, the proposed method was compared with Berger and greedy algorithms. The simulation results indicate that the proposed algorithm reduced the makespan and executive cost when tasks were added. It also increased fairness and load balancing. Moreover, it made the optimal use of machines possible and increased user satisfaction. According to evaluations, the proposed algorithm improved the makespan by 80%.
@artical{f642017ijcatr06041005,
Title = "Cost-Efficient Task Scheduling with Ant Colony Algorithm for Executing Large Programs In Cloud Computing",
Journal ="International Journal of Computer Applications Technology and Research(IJCATR)",
Volume = "6",
Issue ="4",
Pages ="172 - 212",
Year = "2017",
Authors ="Fatemeh Imani , Shiva Razzaghzadeh , Masoud Bekravi"}
  • The scheduling algorithm is regarded as one of the most important challenges and problems in the cloud
  • The tasks sent by the user are classified and prioritized with respect to the quality parameters in the computing unit
  • The ant colony optimization (ACO) algorithm was adapted from social theories with a fair and accurate resource allocation approach based on machine performance and capacity
  • The virtual machine is freed after task processing is done