FAT-ETO: Fuzzy-AHP-TOPSIS-Based Efficient Task Offloading Algorithm for Scientific Workflows in Heterogeneous Fog–Cloud Environment

Abstract

Due to the inherent variation across fog, cloud and end devices, task offloading and resource allocation have become complicated issues in a heterogeneous fog–cloud computing environment (HFCE). This study makes an effort to resolve the issue using two popular multi-criteria decision-making (MCDM) techniques, i.e. the analytic hierarchy process (AHP) and technique for order of preference by similarity to ideal solution (TOPSIS). In this study, we present a rank-based computation offloading algorithm for mapping of workflow tasks on three tiers of resources in HFCE. The five performance criteria taken into account are execution time, energy consumption, total cost, resource availability at a specific tier and the processing speed of resources. These performance criteria of the fog, cloud and end devices are evaluated by cascading two separate MCDM techniques. The AHP is used to calculate the priority weights between all the five criteria of evaluation. The TOPSIS method is used to rank the final fog, cloud and end devices, based on the weights of criteria yielded by AHP, and to calculate fuzzy values before offloading each corresponding task. Afterwards, a task can be offloaded to a corresponding resource tier based on the rank. Simulation results demonstrate that the proposed algorithm outperforms conventional offloading algorithms in the HFCE by including performance and cost criteria while offloading decision making, especially in cases where the amount of tasks is large.