Cloud computing is an important step in making distributed computing widely usable, with great potential for growth. One key service in cloud computing is infrastructure as a service (IaaS), which uses virtualization technology to combine resources from different locations into a single pool for easier management and use. IaaS provides these resources as virtual machines (VMs).To make the best use of resources, reduce costs, and save computing time, we need to optimize how VMs are allocated. This paper explores various scheduling algorithms, including first-come, first-serve (FCFS), round-robin scheduling (RRS), and genetic algorithm (GA), to achieve balanced workload distribution across virtual machines. However, traditional algorithms such as FCFS and RRS face limitations in handling dynamic workloads, resource heterogeneity, and unpredictable task execution times, often leading to inefficient resource utilization and increased processing delays. To address these challenges, this study integrates a genetic algorithm (GA) workload distribution approach within these frameworks to improve their adaptability. The performance of these algorithms is evaluated using the NetBeans simulator under various conditions, with metrics including task completion time, resource utilization, cost efficiency, and RAM speed. The findings reveal that the GA outperforms both FCFS and RRS, min–max offering superior scheduling efficiency, more effective workload distribution, and better resource management, resulting in enhanced service quality and operational performance in cloud data centers. This research highlights the potential of optimized scheduling techniques in overcoming traditional algorithm limitations while laying the groundwork for further advancements in intelligent resource management.