Multi‐objective task scheduling in cloud computing

Abstract

Summary Cloud computing services are used to fulfill user requests, often expressed in the form of tasks and their execution in such environments requires efficient scheduling strategies that take into account both algorithmic and architectural characteristics. Unfortunately, this problem is known to be NP‐hard in its general form. Despite the fact that several studies have been published in the literature, there are still interesting and relevant questions to be addressed. Indeed, most of the previous studies focus on a single objective and in the case where they deal with a set of objectives, they use a simple compromise function and do not consider how each of the parameters might influence the others. To this end, we propose an efficient task scheduling algorithm which is based on the pollination behavior of flowers and makes use of both Pareto optimality principle and TOPSIS technique to improve the quality of the obtained solutions. Both single and multiobjective optimization variants are investigated. In the latter case, three optimization criteria are considered, namely, minimizing the time makespan or schedule length, the execution cost, and maximizing the overall reliability of the task mapping. Different test‐bed scenarios and QoS metrics were considered and the obtained results corroborate the merits of the proposed algorithm.

Modified Artificial Bee Colony Algorithm for Load Balancing in Cloud Computing Environments

“…Malti, et al [34] offer a highly effective task scheduling method that leverages flower pollination behavior. This algorithm incorporates the Pareto optimality principle and the TOPSIS approach to enhance the quality of the solutions achieved.…”

An Efficient Combination of Genetic Algorithm and Particle Swarm Optimization for Scheduling Data-Intensive Tasks in Heterogeneous Cloud Computing

“…Pirozmand et al [19] and Pradhan et al [20] exploited GA to optimize the energy consumption and the makespan for scheduling tasks in clouds. Malti et al [21] proposed an efficient task-scheduling algorithm based on a flower pollination algorithm to improve makespan, resource cost, and execution reliability. Two works proposed by Mangalampalli et al [22,23] used cat swarm optimization (CWO) and whale optimization algorithm (WOA) to schedule tasks on virtual servers for cloud computing, respectively, to improve makespan and energy consumption.…”