Multi-Objective Local Pollination-Based Gray Wolf Optimizer for Task Scheduling Heterogeneous Cloud Environment

Abstract

The rebel of global networked resource is Cloud computing and it shared the data to the users easily. With the widespread availability of network technologies, the user requests increase day by day. Nowadays, the foremost complication in Cloud technology is task scheduling. The cargo position and arrangement of the tasks are the two important parameters in the Cloud domain, which can provide the Quality of Service (QoS). In this paper, we formulated the optimal minimization of makespan and energy consumption in task scheduling using Local Pollination-based Gray Wolf Optimizer (LPGWO) algorithm. In the hybrid concept, Gray Wolf Optimizer (GWO) algorithm and Flower Pollination Algorithm (FPA) are combined and used. In the presence of GWO, the best searching factor is used to increase the convergence speed and FPA is used to distribute the data to the next packet of candidate solution using local pollination concept. Chaotic mapping and OBL are used to provide a suitable initial candidate for task solutions. Therefore, the experiments delivered better task scheduling results in low and high heterogeneities of physical machines. Ultimately, the comparison with the simulation results had shown the minimum convergence speed of makespan and energy consumption.