Optimizing Cloud Task Scheduling Through Innovative Metaheuristic Algorithm and Impulsive Fuzzy C-Means

Addressing challenges in cloud computing infrastructure scheduling, such as computation time, budget constraints, and load balancing, poses inherent difficulties. Various task scheduling techniques, including Genetic Algorithms and Honey Bee Foraging, have been introduced to enhance cloud data center performance across diverse scheduling criteria. The NP-hard (non-deterministic polynomial-time hardness) nature of the task scheduling problem arises from the linear increase in solution combinations concerning the issue’s scale, incorporating factors like the number of tasks and computer resources. This complexity makes the efficient arrangement of user responsibilities a challenging endeavor. This study propose metaheuristics tailored for cloud computing and a load-balanced job scheduling approach based on clustering. The suggested algorithm, IGFCM-EDQL, employs a credits-based task scheduling strategy to minimize makespan, maximize resource utilization, and dynamically reduce SLA violations by clustering incoming jobs onto available Virtual Machines (VMs) in a load-balanced fashion. Real-time criteria, namely Task-Length, Makespan, Task-Priority, Deadline, Degree of Inequality, and Cost, are utilized to categorize cloudlets and virtual machines. The performance of proposed task scheduling algorithm is evaluated using contemporary methodologies for task scheduling, and the results are presented in this study.