Edge/cloud task offloading for time-sensitive applications: a hybrid approach with ant colony optimization and simulated annealing

• Main Author: Ramasubramanian, Shreya
• Other Authors: Arvind Easwaran
• Format: Final Year Project
• Language: English
• Published: Nanyang Technological University 2023
• Subjects: Engineering::Computer science and engineering::Computer systems organization::Performance of systems; Engineering::Computer science and engineering::Computer systems organization::Computer-communication networks; Engineering::Computer science and engineering::Computer systems organization::Computer system implementation
• Online Access: https://hdl.handle.net/10356/171977
• Institution: Nanyang Technological University

With the proliferation of Internet of Things (IoT) devices, the volume of big data generated and the number of IoT-powered applications have surged. Processing this data with low latency and minimal data transport costs while upholding Quality of Service (QoS) standards has become imperative. Fog computing has emerged as a solution, introducing computational nodes at the edge of the network to reduce reliance on the centralized cloud. While fog computing enhances performance, it also introduces complexity by adding limited, dynamic, and heterogeneous resources to the network. This paper explores the integration of Ant Colony Optimization (ACO) algorithms with Simulated Annealing, a local search algorithm, to expedite the scheduling of deadline-sensitive task offloading in fog computing environments. The ACO algorithm has demonstrated its capability to provide close to optimal allocation solutions, yet it can converge to a local optimum, giving less successful allocations. Simulated Annealing offers a potential solution by allowing the algorithm to explore suboptimal solutions, fostering the discovery of higher-quality solutions more rapidly. Through this investigation, the paper aims to determine whether combining ACO with Simulated Annealing can enhance the efficiency of task offloading in fog computing scenarios, enabling the fulfilment of QoS standards while mitigating latency and data transport costs.