Multi‐objectivetask scheduling in fog computing using improved gaining sharing knowledge based algorithm

Abstract

Summary In the modern Internet of Things (IoT) era, several applications generate a vast amount of data and that needs to be handled appropriately. The conventional cloud computing system delivers us with enormous resources to manage such voluminous data. Despite that, the growing demands of IoT applications on minimal energy consumption, minimal latency, the privacy of data, data processing based on location, and maximum Quality of Service(QoS) impels the advent of fog computing. As the devices in the fog layer are heterogeneous, distributed, and resource‐constrained, how the fog resources are effectively utilized for executing latency and time‐sensitive applications is a primary challenge. The data generated by the IoT devices are voluminous and produce overhead in network bandwidth during transmission and slow down the response time. This article addresses the task scheduling problem in fog computing to minimize the makespan and energy consumption. The proposed work consists of two phases. In the first phase, the tasks are ordered based on the heuristic method, heterogeneous earliest finish time (HEFT). Then the ordered tasks are scheduled by applying the improved gaining sharing knowledge (IGSK) based algorithm. To reduce the energy consumption dynamic voltage frequency scaling (DVFS) is applied in the proposed scheme. To evaluate the proposed scheduling method the simulations are performed on different workflows with varying sizes. The performance evaluation results exhibit that the proposed work outperforms the task scheduling with similar methods in terms of makespan and energy consumption.

Opposition‐based improved memetic algorithm for placement of concurrent Internet of Things applications in fog computing

“…The combined optimization of service caching placement and offloading of the computation‐intensive task is proposed in Reference 20. The authors in reference 21 and reference 22 proposed an improved gaining sharing knowledge based optimization algorithm and entropy based complex proportional assessement for efficient application placement in fog computing.…”

RL-JSO: A Hybrid Q-Learning and Jellyfish Search Optimizer for Task Scheduling in Smart Homes Using a Fog-Assisted Cloud Architecture

“…The paper [24] The paper [25] resolved the challenges of task scheduling in fog model to reduce makespan and energy consumption. It proposed a two-phase approach involving the "Heterogeneous Earliest Finish Time (HEFT)" heuristic for task ordering and the "Improved Gaining Sharing Knowledge (IGSK)" algorithm for task scheduling.…”

“…• Not always necessary to use sophisticated schedulers [25] • Task scheduling in fog computing using improved gaining sharing knowledge algorithm.…”

Entropy‐based complex proportional assessment for efficient task scheduling in fog computing

“…The arithmetic optimization algorithm 43 is used in IoT workflow scheduling to minimize makespan and energy. The authors in Reference 44 proposed a multi–objective task scheduling algorithm based on improved gaining sharing knowledge‐based optimization to minimize the makespan and energy consumption.…”

“…The datasets are illustrated in Figure 5 and the characteristics of these workflows are given in Table 13. From the astronomical images the Montage workflow 44 generates the mosaic, the LIGO workflow is used to investigate gravitational waves created by various occurrences in the universe and the Epigenomics 45 is used to determine the epigenetic status of a human cell during large scale genome. Similarly, the Cybershake 46 is used to describe the risk of earthquakes in a given area by generating a seismic hazard curve.…”