Multi-Objective Security Driven Job Scheduling for Computational Cloud Systems

This paper proposes a multi-objective parallel job scheduling algorithm for a Computational Cloud environment. We present a fault-tolerant, scalable and efficient solution 1:4 optimizing scheduling of N independent jobs on 114 parallel machines that minimizes two objectives simultaneously, namely the failure probability and the total completion time of all the jobs. Obtaining an optimal solution for this type of complex, large-sized problem in a reasonable computational time using traditional approaches or optimization tools is extremely difficult. As this problem is NP-hard in the strong sense, a meta-heuristic method which is the second version of the non-dominated sorting genetic algorithm (NSGA-II) is proposed to solve this problem. This approach is based on the Pareto dominance relationship, providing no single optimal solution, but a set of solutions which are not dominated by each other. The performance of the presented model and the applied GA is verified by a number of numerical experiments. The related results show the effectiveness of the proposed model and GA for small and medium-sized problems.