The paper proposes a resources allocation method for EON based on a modified RDM using squatting and kicking strategies in order to prioritise the usage of channels and enhancing the total utilisation. It can also be used as an admission control. The mechanism is described and compared to existing proposals, regarding the flexibility in\ud the allocation per class. The strategy regards the EON constraints and assumes the optical spectrum to be partitioned and reserved to several different classes. These partitions are allocated according to the priority of the demands and the strategy proposed. The performance metrics are explained and analysed via demands and resources allocation examples.Peer ReviewedPostprint (published version

The virtual circuit characteristic of MPLS and GMPLS architectures has enabled the implement of Traffic Engineering (TE) and Differentiated Services (Diff-Serv) profile to networks. Both schemes are possible by the application of preemption policies and priority of path assignment. With the advent of Path Computation Element (PCE) the routing task can be centralized and properly (re-)designed to accommodate MPLS' and GMPLS' Diff-Serv and TE characteristics. However, as a centralized system, the complex characteristics of heterogeneous networks and the extra data involved on TE and Diff-Serv constrains and requirements for path computations may overload the PCEs' processing capacity and radically increase the number of PCEP control messages exchanged, jeopardizing the scalability of the system in general. In this article we present an analysis on the current standardized PCE tools in a WDM network in terms of the number of control messages exchanged impact on the system and its optimization performance. INTRODUCTIONAs networking advance technically the variety and amount of requirements service providers deal with proportionally increases. In such a scenario Traffic Engineering (TE) mechanisms may be used by service providers as an efficient tool for managing and balancing resources throughout a network and also as an assistance to provide Quality of Service (QoS) adequately conforming to Service Level Agreements (SLA) [1].One of the diverse requirements that TE helps attend is the varied performance guarantees demanded by multimedia applications. Due to their intrinsic virtual circuit switched characteristics MPLS and GMPLS networks support routing paths to be specified and therefore qualified according to their ability to satisfy diverse QoS requirements. Differentiated Service (Diff-Serv) architecture permits traffic flows to be classified into classes and managed following specific rules so that higher priority traffic can receive preferential service [2].Diff-Serv aggregated to MPLS networks aid advanced planning procedures in order to provide specific QoS guarantees [2]. This combination, of Diff-Serv and MPLS resulted in the Diff-Serv aware TE (DS-TE), a scalable solution to support QoS and efficiently balanced network's use. The DS-TE is implemented in MPLS and GMPLS network through distributed constrained path computation.In heterogeneous multi-domain and multi-layer networks, however, constrained-based path computation becomes extremely complex and resource invoking. Path Computation Element (PCE) architecture was developed as a possible solution to this issue [3]. In PCE architecture dedicated devices centralizes the function of constrained based path computation releasing the burden from the MPLS or GMPLS nodes. Furthermore PCEs in different layers or domains may collaborate with each other to perform an optimal end-to-end path computation.In the PCE architecture path computation is no longer responsibility of MPLS or GMPLS nodes, therefore, the DS-TE features of virtual circuit switched networks had t...

Differentiated services provision according to clients or applications is an added feature that network operators ambition to fully provide. One of the possible approaches is to map DiffServ into MPLS based network. However in an IP over WDM (Wavelength Division Multiplexing) network data flow traverse multiple layers and, therefore, Class-Type information from the MPLS DiffServ layer is eventually lost. Besides, with the implement of a Path Computation Element (PCE) on multi-layer and multi-domain networks, the bandwidth\ud constraints used to differentiate services are no longer applied. This paper presents a proposal for a QoS DiffServ approach applied on PCE as a traffic grooming technique for WDM core networks and the problematic involved on the application of this technology on multilayer and multi-domain networks.Postprint (published version

Description of online digital content is currently extremely necessary to facilitate a diverse amount of resource sharing over the internet. Many times, content is shared and reused within a virtual community. Virtual communities tend to have their own specific needs of resources, and tend to use a specific vocabulary to describe content. Members of virtual communities also tend to have specific motivations for participating and sharing information and knowledge with other members. In this chapter, the authors discuss the benefits of community members generating content description by analyzing the current literature on the matter. Then, the authors present two studies they have held where they assess the metadata generated by users of an IPTV system and by members of two different virtual communities.