Optimizing Workflow Scheduling in Cloud Computing Through Comparative Study of Dynamic Group and Prioritize Scheduling (DGPS) and Traditional Algorithms

Abstract

In cloud computing environments, efficient workflow scheduling is critical for optimizing resource utilization and minimizing response times. This study introduces and evaluates a new scheduling algorithm—Dynamic Group and Prioritize Scheduling (DGPS)—and compares its performance with three traditional algorithms: First-Come-First-Served (FCFS), Shortest Job First (SJF), and Round Robin (RR). The DGPS algorithm dynamically groups tasks based on their attributes and prioritizes them before allocation to Virtual Machines (VMs), aiming to enhance scheduling efficiency. Through simulations with 5 VMs and 50 tasks, the performance metrics of average response time and standard deviation were analyzed. The results indicate that DGPS provides a balanced performance with stable response times, while SJF achieves the lowest average response time but with moderate variability. FCFS offers slightly better response times than DGPS but with higher variability, and RR demonstrates the highest response times and standard deviations. This research highlights the effectiveness of DGPS in achieving consistent and efficient task scheduling in cloud environments.