Grand scale scheduling for admission control

Admission control and resource scheduling govern and manage the usage of resource within a video streaming proxy server coping to strive for a better quality of service. Such resources include network bandwidth, disk bandwidth, processing power and memory. Inside the proxy server, these usages are scheduled during admission control upon new request acceptance. Scheduling schemes such as constant-data-length (CDL) and constant-time-length (CTL) seeking to improve these resources' utilization have been proposed. However they focused mainly on an individual requirement alone and this leads to inefficiency. In this paper, through an analysis of the relationship of resource utilization between individual schedules and their aggregate, we can determine the worst case requirements. Based on this relationship, we propose the grand scale scheduling which can provide a substantial utilization improvement.