Telecom operators are starting the deployment of Content Delivery Networks (CDN) to better control and manage video contents injected into the network. Cache nodes placed close to end users can manage contents and adapt them to users' devices, while reducing video traffic in the core. By adopting the standardized MPEG-DASH technique, video contents can be delivered over HTTP. Thus, HTTP servers can be used to serve contents, while packagers running as software can prepare live contents. This paves the way for virtualizing the CDN function. In this paper, a CDN manager is proposed to adapt the virtualized CDN function to current and future demand. A Big Data architecture, fulfilling the ETSI NFV guide lines, allows controlling virtualized components while collecting and pre-processing data. Optimization problems minimize CDN costs while ensuring the highest quality. Re-optimization is triggered based on threshold violations; data stream mining sketches transform collected into modeled data and statistical linear regression and machine learning techniques are proposed to produce estimation of future scenarios. Exhaustive simulation over a realistic scenario reveals remarkable costs reduction by dynamically reconfiguring the CDN.Peer ReviewedPostprint (author's final draft

No abstract

Abstract-Management is a critical issue in optical networks. Highly impacting in management, routing in optical networks, is often decoupled into both the path selection and the wavelength assignment problems. In the former, shortest path based approaches are often applied to compute paths. Although such approaches offer advantages in terms of network loading, they are restricted by the fact, that routes are selected without taking into account parameters pertaining to any explicit and desired service guarantees. Thus, the wavelength assignment process may assign a non-optimal wavelength with respect to incoming traffic requirements and any associated quality of service specifications. This paper addresses the lightpath selection problem by proposing a novel route selection scheme where routes are determined based upon the twin criteria of minimizing the number of hops and balancing the network load, resulting in a reduction in both network congestion and blocking probability.

The ongoing proliferation of new services, applications, and contents is leading the Internet to an architectural crisis owing to its inability to provide efficient solutions to new requirements. Clean-slate architectures for the future Internet offer a new approach to tackle current and future challenges. This proposal introduces a novel clean-slate architecture in which the TCP/IP protocol stack is decoupled in basic functionalities, that is, atomic services (ASs). A negotiation protocol, which enables contextaware service discovery for providing adapted communications, is also specified. Then, we present how ASs can be discovered and composed according to requesters' requirements. In addition, a media service provisioning use case shows the benefits of our framework. Finally, a proof-of-concept implementation of the framework is described and analyzed. This paper describes the first clean-slate architecture aligned with the work done within the ISO/IEC Future Network working group.

Many carriers and Service Providers (SPs) use different network domains. The Path Computation Element MPLS to achieve Traffic Engineering (TE) objectives, such as (PCE) architecture has been proposed by the IETF [2], based network resources optimization, support for end-to-end QoS on the idea of decoupling Path Computation from Path guarantees services, aggregated traffic measurement and fast Establishment processes, taking the former out of the routing recovery against link/node/Shared Risk Link Group (SRLG) X ,failures. The three main Traffic Engineering components of hardware, while keeping the latter under Control Plane MPLS are the routing component, responsible for the discovery responsibility. The PCE functionality is orthogonal to the of the network topology, the path computation component, forwarding capacity of the node in which it is implemented. responsible for the assignment of resources to traffic demands, Centralized, off-line route computation has been practised and the signaling component, responsible for the establishment of in transport networks since long ago. Given the network the Label Switched Paths (LSPs) along the computed path. Path Computation under QoS and administrative constraints is atopology and a staic traffic matrix a route server can compute usually performed sub-optimally in network nodes, which are manaotmal dribut resources.antheetyp ao mnewrs mainly dedicated to traffic forwarding. Off-load of this task to management procedures (either automatic and/or manual)