A Hybrid Task Scheduling Algorithm for Improving Resource Utilization in Cloud Environments

Cloud computing has emerged as a fundamen­tal technology that enables scalable and on-demand access to computing resources such as storage, processing power, and networking. However, efficient task scheduling remains one of the major challenges in cloud environments due to the heterogeneous nature of resources and dynamic workload conditions. Ineffi­cient scheduling can lead to poor resource utilization, increased execution time, and higher operational costs. To address these issues, this study proposes a hybrid task scheduling algorithm designed to improve resource utilization and optimize task exe­cution in cloud computing environments. The proposed approach integrates heuristic scheduling techniques with machine learning-based optimization to dynamically allocate tasks to available virtual machines based on resource availability, workload char­acteristics, and execution priorities. The algorithm aims to minimize makespan, improve through­put, and balance workload across computing nodes. Experimental evaluation is conducted using a simulated cloud environment with multiple datasets representing heterogeneous workloads. The results demonstrate that the proposed hybrid scheduling model significantly improves resource utilization and reduces task completion time compared to traditional scheduling methods such as First Come First Serve (FCFS), Round Robin, and Min-Min algorithms. The proposed framework also shows improved load balancing and scalability under varying workload conditions. These findings suggest that hybrid scheduling techniques can effectively enhance the efficiency and performance of cloud computing infrastructures. Future research may explore integra­tion with reinforcement learning models and real-time workload prediction mechanisms to further optimize scheduling decisions.