Research Article
HGA-RRHC: Hybrid Genetic Algorithm Random- Restart Hill-Climbing Dynamic Task Scheduling in Edge-Cloud Computing
@INPROCEEDINGS{10.4108/eai.28-4-2025.2358078, author={Panchagnula Kamakshi Thai and Shanker Chandre}, title={HGA-RRHC: Hybrid Genetic Algorithm Random- Restart Hill-Climbing Dynamic Task Scheduling in Edge-Cloud Computing}, proceedings={Proceedings of the 4th International Conference on Information Technology, Civil Innovation, Science, and Management, ICITSM 2025, 28-29 April 2025, Tiruchengode, Tamil Nadu, India, Part II}, publisher={EAI}, proceedings_a={ICITSM PART II}, year={2025}, month={10}, keywords={cloud computing edge computing task scheduling hill-climbing genetic algorithm}, doi={10.4108/eai.28-4-2025.2358078} }- Panchagnula Kamakshi Thai
Shanker Chandre
Year: 2025
HGA-RRHC: Hybrid Genetic Algorithm Random- Restart Hill-Climbing Dynamic Task Scheduling in Edge-Cloud Computing
ICITSM PART II
EAI
DOI: 10.4108/eai.28-4-2025.2358078
Abstract
Internet of Things (IoT) grows increasingly diverse, new, challenging, computationally complex, and time-sensitive as more and more devices connect to the internet. Applications like object detection, smart homes, and smart grids have emerged. Nevertheless, more conventional architecture in cloud computing raises the problem of high latency, which would not fit IoT devices due to their restricted processing and storage power. This is solved by edge computing because the edge devices are deployed close to IoT devices, offering low-latency computation capability. This paper introduces a new hybrid method called HGA-RRHC for dynamic task scheduling in IoT Edge-Cloud environments. The method aims to address the previously mentioned issues. To incorporate this, the system permits the user to choose from different artificial intelligence (AI) approaches and define the number of tasks and nodes to schedule. Each task means a randomly chosen deadline and necessary computational power; each node is randomized given speeds and costs. The applied AI methods are the Hill-climbing algorithm, the Random Restart Hill-climbing (RRHC), and the Genetic Algorithm (GA). This proposed HGA-RRHC method capitalizes on the global searching capability of GA and the cellular automata-based neighborhood programming for task-node assignment.
