Abstract
For the typical optimal problem of task scheduling in cloud computing, this paper proposes a novel resource scheduling algorithm based on Social Learning Optimization Algorithm (SLO). SLO is a new swarm intelligence algorithm which is proposed by simulating the evolution process of human intelligence and has better optimization mechanism and optimization performance. This paper proposes two learning operators for task scheduling in cloud computing after analyzing the characteristics of the problem of task scheduling; then, by introducing the Small Position Value (SPV) method, the two learning operators with continuous nature essence are enabled to solve the problem of task scheduling, and then the improved SLO is employed to solve the problem of cloud resource optimal scheduling. Finally, the performance of improved SLO is compared with existing research work on the CloudSim platform. Experimental results show that the approach proposed in this paper has better global optimization ability and convergence speed.
Highlights
Cloud computing is a business service model and computing model which integrates grid computing, parallel computing and P2P technologies[1]
For solving task scheduling problem efficiently, this work designs some optimization operators for different spaces of Social Learning Optimization Algorithm (SLO), Small Position Value (SPV) method is used to discrete the individuals of improved SLO, so as to enable the improved SLO could solve the discrete task scheduling problem
The performance of improved SLO is compared with other methods on the CloudSim platform
Summary
Cloud computing is a business service model and computing model which integrates grid computing, parallel computing and P2P technologies[1] It is a pay-as-you-go model which provides resources at lower costs with greater reliability. Above evolutionary algorithms still have some shortcomings, such as slow convergence speed, easy to fall into local optimum, etc These shortcomings reduce the efficiency of task scheduling problem solving. To solve the task scheduling problem efficiently, this paper proposes a new approach based on improved Social Learning Optimization Algorithm (SLO) [17]. SLO is a new evolutionary algorithm which is proposed by simulating the evolution process of human intelligence, it consists of three co-evolution spaces: micro-space, learning space and belief space.
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
