In this paper, the performance scalability in clusterbased Video-On-Demand (VOD) servers is studied with several grouping configurations of cluster nodes. To find performance bottlenecks of VOD servers, the monitoring functions are employed and the maximum Quality of Service (QoS) streams are measured under the various requests including VCR functions. From detailed experimental results, a new admission control method is proposed, that is based on available system resources and the actual amount of resource consumed for QoS streams. The proposed method provides not only the more scalable QoS in cluster-based VOD servers then legacies but also the enhancement of resource utilization by guaranteeing the maximum number of QoS streams than legacies.