Network Management Architecture:
  • Clark, Partridge, Ramming, and Wrocolawski presented at SIGCOMM 2003 A Knowledge Plane for the Internet making a case for a new network management architecture, that utilizes machine learning and cognitive systems to take observations, asserations, and explanations and provide reasoning to make changes to the system, all of this in the presence of incomplete, inconsistent, and potentially misleading information.

  • Peterson et. al. have proposed an Information Plane architecture they prototyped and a paper on a monitoring application called PlanetSeer. PlanetSeer uses passive monitoring on PlanetLab nodes to watch for anomalies then actively probes anomalies from a distributed subset of nodes. The analysis of the faults is an offline process. The Information Plane was inspired by some of the same talks as The Knowledge Plane, and Peterson draws the distinction that the Info Plane could provide some of the functions for the Knowledge Plane, but that his research does not include any of the cognitive or machine learning aspects of the complete Knowledge Plane as Clark envisions.

  • Sterritt, Marshall, Shapcott, and McClean in IEEE Int. Conf. Systems Man and Cybernetics 2000, Exploring Dynamic Bayesian Belief Networks For Intelligent Fault Management Systems. They present a nice introduction to Bayesian Belief Networks, Dynamic Bayesian Belief Networks, and how they may be used for many different network management functions.

  • RMP: RECONFIGURATION MANAGEMENT PLANE FOR THE SUPPORT OF POLICY BASED NETWORK RECONFIGURATION by N. Alonistioti, A. Glentis, F. Foukalas, A. Kaloxylos, and A Scalable Architecture for Network Fault Diagnosis in the Knowledge Planeby Lee, Faratin, and Bauer

  • ITU-T's TMN Telecommunications Management Network - the classic telco network management model, a tutorial is available here. FCAPS describes the functionality required to implement the TMN.

Fault Localization:
Network Management Tools: