Developing IP applications is important but troublesome in practice since reusable components are usually system-call-level service routines. It is becoming more and more indispensable in this network generation. Under the concept that an IP application consists of distributed objects communicating with one another, the authors define an architecture and design a framework for IP applications. The architecture is a 4-layered model whose layers are communication, ORB (Object Request Broker), state transition, and application. The communication layer is for implementing virtual circuit and datagram communication. The ORB layer is a platform for message exchange over a network. The state transition layer packs automata extracted from a client and a server. The application layer includes objects handling the client user interface and server database. Based an the architecture, they designed a framework for the state transition and application layers. The framework for the state transition layer implements client and server automata. The user interface is categorized into three types: OS command, subcommand, and browser, and then a framework for each type is designed. A database framework provides mechanisms to serve its object-oriented interface. Through the implementation of several IP applications using a prototype version of the framework, usefulness of the architecture and the framework is demonstrated.

The performance of a broadcast star network is analyzed, under the assumption of synchronous operation of a network. An exact analysis for a broadcast star network with an infinite station population, an exact analysis for a small network with a finite station population, and an approximate analysis for a large network with a finite station population are presented. In synchronous operation, the channel time is slotted, and stations transmit only at the beginning of a slot. The throughput and the distribution of transmission delays are obtained. It is shown through simulations that a broadcast star operating under synchronous mode yields better performance than when operating under asynchronous mode, where transmissions of packets are not confined to the beginning of slots and stations start transmission at any time.>

The design is presented of a broadcast star network that uses a collision-avoidance switch. This network combines the benefits of random access (low delay when traffic is light; simple, distributed, and therefore robust protocols) with excellent network utilization. Performance analyses of a slotted broadcast star network are then presented to obtain distributions of the transmission delay: an exact analysis for a network with an infinite number of stations, and an approximate analysis for a network with a finite number of stations.>