Abstract:

Workflow scheduling is one of the critical problems in cloud environment. Lots of strategies are presented to solve it. However, most of them only care about satisfying the time and cost QoS requirements of the users, such as the deadlines of the tasks, and cannot guarantee the security of scheduling. But the security is one of the most concerned issues for the cloud users. To solve this problem, a security constraint model for workflow scheduling problem is presented. Based on it, the Variable Neighborhood Particle Swarm Optimization(VNPSO) is adapted and implemented as the scheduling algorithm. Two representative metaheuristic based scheduling algorithms including MaxMin Ant Colony Optimization (MMACO) and Genetic algorithm (GA) are also implemented as the candidate scheduling algorithms. CloudSim is set up as the simulation platform, and the simulation experiments show that the variable neighborhood search heuristic provides a good balance between global exploration and local exploitation and makes VNPSO feasible and effective for scheduling workflow tasks.

Key words: cloud computing;resource scheduling;workflow;security constraints;VNPSO