A resource reservation and scheduling algorithm for learning on-demand system over satellite and cable network

Distance learning systems over satellite and ground wired network provide a multimedia study environment and unlimited learning opportunities. Learning on demand has become an efficient and popular learning mode. Many learning on demand systems based on IP networks have been implemented and applied, such as real system, media server and IBM Learning Space, etc. However, they neither meet a user´s reservation request for education resources, nor can they provide a resource optimization scheduling scheme when many people wish to play the same kind of education resources at the same time. In this paper, we propose a new and practical resource reservation and scheduling algorithm for learning on demand based on satellite and ground IP network (RODS for short). In RODS, users provide a request for resources to the server station via ground Internet, then RODS employs the improved maximal queue length (IMQL) algorithm to schedule reservation requests and then generate an optimized sending scheme of multimedia courseware. RODS sends the corresponding resources to remote sites via satellite transmission. The prototype of RODS has already been implemented and tested on the platforms of satellite and CERNET (China Educational and Research Network), and the algorithm was proved to be efficient, flexible and practical in supporting learning on demand.