| Last Updated:
Creating DOI. Please wait...
Computing systems, in general, are becoming more complex, heterogeneous and wide range as, for example, in the context of the Internet of Things (IoT), Future Internet (FI), Cloud Computing and Smart Cities, among several other ongoing trends. These new computer systems typically have as a characteristic a very large number of users, possibly mobile, with a support of heterogeneous networks and a great variety of requirements and dynamic traffic profiles where the NP-hard resource allocation problem is one of the focuses of research. For these new computer systems, including its specific resource allocation method, it is typically required solutions that are dynamic, autonomic, integrated, secure and, in some cases, federated. Virtualization, autonomic computing, software defined networking (SDN), cloud computing and network federation are examples of emerging and evolutionary paradigms and enablers in a challenging context of research and development. This research project investigates a new set of solutions for the resource allocation problem that allows computing dynamic solutions with autonomic features and feasible computation time (possibly on-the-fly) in this new computer network context and scenario.
In order to achieve this goal, the RePAF project (Resource Provisioning and Allocation Framework) proposes the development of a new integrated model for provisioning and allocation of resources (RPAM - Resource Provisioning and Allocation Model) and a framework (FDARM - Framework for Dynamic and Autonomic Resource Management) integrating the functions of RPAM model aimed, in short, the provisioning and dynamic allocation of resources with autonomic characteristics. The F-DARM framework, based on the resource allocation model RPAM, adopts the paradigm SDN / OpenFlow as part of its architecture, integrates autonomic features and adopts a strategy of acquiring knowledge using learning techniques for computing dynamic resource allocation solutions with autonomic features.
Regarding the focus of the research, the proposed solution supports resource allocation for MPLS network, Elastic Optical Networks (EON), IoT and Network Function Virtualization (NFV).