In this paper, we argue that the standard-based Intelligent Agents constitute an attractive technology to address the challenging issues arising from the unpredictable nature and dynamism of modem heterogeneous networks.Paramount among these issues are diverse services, networks and technologes, multiple vendor equipment, loosely organized management applications and inconsistent databases, multiple management protocols and multiple data representations.The term agent is highly overused. In this paper, an agent is defined to bea computational entity, which acts on behalf of others, is autonomous, is both proactive and reactive, and exhibits certain degree of capabilities to learn, cooperate and move. There are currently two standards regarding agent-based systems: Foundation for Intelligent Physical Agents (FIPA) and Object Management Group's (OMG) Mobile Agent System Interoperability Facilities (MASIF). An OMG work group has been created recently with a mandate to merge the two standards with proposals from a number of other contributors. The authors are actively involved in an effort to deliver to Lucent's business units a standardabiding agent framework called Lucent Intelligent Network Agents, or LucINA for short.A classical clientkerver (managedagent) approach to network management can be implemented with agents. Such an approach has a number of critical advantages over traditional implementations, which are discussed in the paper in detail using a comparative analysis. In an agent-based implementation, a pair of agents acts as a client (manager) and a server (NM agent). The paper assumes that the agents are FIPA-compliant, and, as such, can communicate using the FIPA Agent Communication Language (ACL). The FIPA-ACL with the standard provisions regarding a language, protocols and ontologies, constitutes a meta-level communication layer, which allows the agents to setup a coherent communication session. In addition to specifying the receiver of the message, the manager identities itself, the ontology with which it proposes to 0-7803-5864-3 0 2000 IEEE
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.