An Intelligent Model for Distributed Systems in Next Generation Networks

Abstract

Abstract. Over the past twenty years, network technology has been improved rapidly in term of speed, performance, component, and functionalities. Therefore, a number of different types of network devices have been developed; this led to an increase in the complexity of network systems. Traditional network structures are inadequate to meet today requirements. It is centralized network which imposes on human operators to have a high experience on how to detect changes, configure new services, recover from failures and maximize Quality of Service (QoS). Therefore, network management involves heavy reliance on experts operators. The adopted centralized network management is not suitable for new technologies emerging, which are complex and difficult to interact among heterogeneous networks that contain different types of services, products and applications from multiple vendors. As a result, the current network management lacks of efficiency and scalability; however, it has an acceptable performance generally. The centralized information model cannot stand and achieve the requirements from such complex, distributed electronic environments. This paper studies the need of distributed systems in next generation networks. Then, the paper presents three network structure paradigms: centralized, hybrid and distributed. After that, Software-Defined Networking (SDN) is described. Finally, the paper proposes a distributed approach for OpenFlow technology using a Distributed Active Information Model (DAIM) which supports an autonomic management of the distributed electronic environment.