Enabling Frameworks for Autonomic Adaptation of Protocols in Future Internet Systems

Abstract

The emergence of several networking standards has been continuous over the last decade. Engineering creativity spawned a wide gamut of innovative technologies for wireless and wireline communications. This increase in technological portfolios, in combination with the requirement to migrate legacy systems and to maximize the use of large investments in network installations, resulted in the design of multiple network evolution paths. Combined with the increasing sophistication of networking technologies, this variety of choice in design has run counter to the simplification – and the efficiency – of management procedures. The task of managing network infrastructures is confronted with an increasingly disharmonious Babel of standards involving interfaces, protocols, topologies, and versions. As a result, there has been a turn of research interest towards an autonomous mode of management where the elements of the managed system display individualistic proactive behavior that strives to maintain their modus operandi within specific bounds. The umbrella term autonomic computing and communications refers to a capability set that includes a system’s ability to monitor selected aspects of its own operation, collect, and record any data resulting from these observations, evaluate its performance under the light of its own operational history, possibly also identifying trends and recurring patterns in the process, and, in the case of subpar performance, undertaking corrective actions targeted to achieve a satisfactory level of performance. To this end, dynamically adaptable protocol stack offer a systemic capacity for change during runtime according to current operational requirements, thus providing an essential framework feature of autonomic systems.