Search

The increased complexity of emerging and future networks and services requires the use of sophisticated but simplified management and control that will minimize the burden of human presence and manual operation. Autonomic network management (ANM), on the one hand, and softwaredefined networking (SDN), on the other hand, have appeared with the promise to meet these requirements and enhance the efficiency of the control and operation of networks. These two emerging technologies share motivation and have confluent goals, e.g., increasing the reliability and efficiency of operation, simplifying network management and control, focusing on means for opening innovation and differentiation for vendors, and aiming at capital expenditures/operating expenditures (CAPEX/OPEX) reduction for providers. This article presents and promotes a framework that extends the G?ANT test bed toward an autonomic OpenFlow (AUTOFLOW) facility, aiming at 1) demystifying the relationship and interplay between ANM and SDN and 2) showcasing that SDN/OpenFlow capabilities can bring customizable ANM into reality. [ABSTRACT FROM PUBLISHER]

Due to the need for supporting diversified applications with vastly expanded use of wireless access, the Future Internet (FI) will set the networks under stress for resources. This motivates the quest for Quality of Service (QoS), through technologies that achieve higher utilization of resources and lower transmission powers and energy consumption. The goat of this paper is to present and validate a solution that achieves these targets, and thus, creates important value-adding opportunities for both the user and the wireless industry. The proposed solution considers the possibility of extending the infrastructure-based Composite Wireless Network (CWN), which is owned by a wireless operator and consists of a set of radio access networks, by network elements and devices that are organized in an infrastructure-less manner. The decision of the CWN to create or not such an infrastructure-less network will be dynamic, should respect traffic, energy and mobility aspects and should also envisage anticipated QoS related benefits for the whole network. Framed within this statement, this paper aims at recommending whether and when the creation of such an infrastructure-less network(s) will be beneficiary for the overall network as an extension to a CWN. For that reason, extensive simulations were conducted so as to cover a large gamut of test cases, thus increasing the validity of the extracted recommendations. Simulation results reveal that there exist specific suitability criteria under which the creation of infrastructure-less networks can be recommended. [ABSTRACT FROM AUTHOR]

Operator-governed opportunistic networks (ONs), which are dynamically created, temporary coordinated extensions of the infrastructure, are the first ingredient for the proposed networking solution. This article presents the concept of cognitive management systems (CMSs) for the management of ONs, their coordination with the infrastructure, and evaluation of an indicative test case as a proof of concept for the aforementioned approach. [ABSTRACT FROM PUBLISHER]