An Multiple Pheromone Algorithm for Cloud Scheduling With Various QOS Requirements
R. Gogulan, A. L. Kavitha · 2012
The Cloud computing is one of the rapidly improving technologies. Cloud computing is a new promising paradigm in distributed and parallel computing. As cloud-based services become more dynamic, resource provisioning becomes more challenging. One of the critical problems in cloud computing is job scheduling because it increases with the size of the grid and becomes difficult to solve effectively. This paper introduces a new algorithm called Multiple Pheromone Algorithm which is belongs to Ant Colony Optimization Algorithm. The objective of MPA algorithm is to dynamically generate an optimal schedule so as to complete the task in minimum period of time as well as utilizing the resources in an efficient way. In this paper three different Quality of Service (QoS) makespan, cost and reliability constraints are considered as performance measure for scheduling. This algorithm is compared with normal Ant colony algorithm, Genetic Algorithm. With the implementation of this approach, the Multiple Pheromone Algorithm (MPA) Algorithm reaches optimal solution as well as obtains the better QoS than ACO and GA.