An Application of Genetic Algorithm to Hierarchical Configuration of ATM Internetworking Domain with a Special Constraint for Scalable Broadcasting

摘要

Presented is a Genetic Algorithm(GA) for dynamic partitioning of an ATM internetworking domain such as LANE(LAN Emulation) that provides guaranteed quality of service for mid-size campus or enterprise networks. There are few researches on the efficient LANE network operations to deal with scalability issues arising from multi-media broadcasting, limiting its potential in multimedia applications. To cope with the scalability issue here, proposed is a decision model named LANE Partitioning Problem(LPP) which aims at partitioning the entire LANE network into multiple virtual LANs. We also demonstrate with some experiments that compared with the current rule-of-thumb practice, the suggested LPP decision model and GA-based solution method significantly enhances performance, thereby providing good scalability in a large internetworking environment such as ATM LANE and mobile ATM LAN.