Network slicing is a key concept in 5G networking. It enables an infrastructure provider (InP) to support heterogeneous services over a common platform by creating a customized slice for each one of them. Once in operation, the slices can be dynamically scaled up/down to match the variation of service requirements. Although an InP generates revenue by accepting a slice request, however it might need to pay a penalty (proportional to the level of service degradation) if a slice cannot be scaled up when required. Hence, it becomes crucial to decide which slice requests should be accepted in order to maximize the net profit of an InP. This paper presents a slice admission strategy based on big data analytics (BDA) predictions. The intuition is to accept a slice request only when it is estimated that no service degradation will take place for both the incoming slice request and the slices already in operation. In this way, the penalty paid by an InP is contained, with beneficial effects on the overall net profit. Apart from simulations, the performance of the proposed admission policy has also been evaluated using emulation. Simulation results show that, in the presence of a high penalty due to service degradation, using BDA predictions brings up to 50.7% increase in profit, as compared to a slice admission policy without BDA. Emulation results for a small network scenario show a profit increase of up to 383% with only a small impact on the slice provisioning time (i.e., due to the processing of BDA predictions).

Connection provisioning in Wavelength Division Multiplexing (WDM) networks needs to account for a number of crucial parameters. On the one hand, operators need to ensure the connection availability requirements defined in Service Level Agreements (SLAs). This is addressed by selecting an appropriate amount of backup resources and recovery strategies for the connections over which services are provisioned. Services requiring less strict availability requirements can be routed over unprotected lightpaths. Services with more strict availability requirements are provisioned over protected lightpaths in order to cope with possible failures in the network. Another important aspect to consider during the provisioning process is energy efficiency. Green strategies leverage on setting network devices in Sleep Mode (SM) or Active Mode (AM) depending on whether or not they are needed to accommodate traffic. However, frequent power state changes introduce thermal fatigue which in turn has a negative effect on the device lifetime. Finally, in multi-period traffic scenarios, it is also important to minimize the number of reconfigurations of lightpaths already established in the network in order to avoid possible traffic disruptions at higher layers. The work presented in this paper tackles the connection provisioning paradigm in an optical backbone network with a multi-period traffic scenario. More specifically the paper looks into the interplay among (i) energy efficiency, (ii) thermal fatigue, and (iii) lightpath reconfiguration aspects. To this end, the Energy and Fatigue Aware Heuristic with Unnecessary Reconfiguration Avoidance (EFAH-URA) is introduced, showing that it is possible to balance the three aspects mentioned above in an efficient way. When compared to the pure energy-aware strategies, EFAH-URA significantly improves the average connection availability for both unprotected and protected connections. On the other hand, it is done at the expense of reduced energy saving.

Being able to account for the value of latency introduced by the transport infrastructure is considered a crucial feature for any 5G service provisioning strategy. This paper proposes a methodology for the automatic monitoring of the link latency in networks with a distributed SDN-based control plane architecture. The proposed link monitoring solution does not introduce any overhead in the OpenFlow channel and in the data plane. Experimental tests demonstrate the accuracy of the proposed methodology and its ability to scale to large network instances.

Background and purposeEffective supervision practices are vital for the educational and professional development of students,for continuous growth of supervisors, as well as for the development of respective scientific fields. In lightof different learning styles (Taylor & Beasley 2005) and having in mind the time resource constraints ofsupervisors, it is not easy to point out the best pedagogical approach to supervision that maximizes thelearning experience. In addition to the traditional individual supervision (IS) style there are other options(e.g., group supervision (GS) and peer supervision (PS)), which offer certain advantages. These threestyles do not exclude each other, but can rather be combined to complement each other’s strengths.In order to maximize the effectiveness of combining multiple approaches, it is essential to understand itsadvantages and disadvantages. Based on a survey of different experiences among supervisors andstudents collected from different Swedish education institutions, our paper suggests ways to optimizethe supervision processes. Moreover, we call it harmonized supervision, and belive that it would savetime and effort for the supervisors, and help students to overcome the individual limitations of eachsupervision style. 

Work done/work in progressIn order to study the preferences of students and supervisors with respect to IS, GS, and PS weconducted a survey among faculty members as well as former students at four higher educationinstitutions (HEIs), where our goal was to aggregate their experiences and learnings. The sampling wasdone in two-stages. First, we selected the HEIs. Due to convenience and connections to specific departments at given HEIs that the authors had, we then sent e-mail invitations to both students andsupervisors at these HEIs. In the second stage, through a voluntary process, respondents from bothgroups took part in the survey. Questions in the survey were inspired by the previous experiences of theauthors, and traditional supervision approaches of the affiliated institutions. We asked informants abouttheir experiences, and what they believed were advantages and disadvantages of each of theexperienced supervision styles. Finally, data was analyzed using descriptive statistics and qualitativeanalysis of open-ended questions. Basically, we looked into which style was used the most and in whichsituations, as well as compared different answers that spoke in favor and against each style.Results/observations/lessons learnedIt is interesting to note that supervisors and students had similar views with respect to IS, GS, and PS. Interms of IS “lack of different perspective” and “limited flows of new idea/opinions” are among thedrawbacks highlighted by both supervisors and students. Interestingly enough, a solution to these issuesis readily available among the benefits of GS and PS, i.e., “New ideas for solving problems” and“Diverse feedback”. This observation leads us to conclude that combining IS, GS, and PS in aharmonized supervision approach. By harmonized supervision we refer to an approach where GS andPS are used as the basis, and where IS is used only when needed.Take-home messageRegardless of the choice of the supervision method, one can note that a mixture of style is moreeffective depending on the learner’s phase, which can be broken down in two main stages. In theinitial phase, the supervisor exercises a more structural and contractual style. For instance, thesupervisor acts as a teacher explaining the research method and the student performs it on a step-bystep basis. The next stage is the training phase, where the supervisor can give the student moreformative assessment support and feedback to develope student's skills until a certain autonomy qualityis achieved. Lastly, the learner becomes a master of the thesis topic and therefore becomes moreindependent. When considering supervision it is important to think about different levels of intellectualdevelopment and the social component of the learning process. At the second phase, i.e. trainingphase, the supervisor can adopt group or peer supervision. Engaging the students in peer and groupsupervision may be conducive to the creation of a more secure learning environment. However, it isessential to provide a constructive group constellation and complementing instructions for peers tomaximize the learning outcomes in an efficient manner.

QoS-based connectivity coordinated by the 5GEx Multi-domain Orchestrator exploiting novel stateful BRPC is demonstrated for the first time over a multi-operator multi-technology transport network within the European 5GEx Sandbox, including Segment Routing and optical domains.

QoS-based connectivity coordinated by the 5GEx Multi-domain Orchestrator exploiting novel stateful BRPC is demonstrated for the first time over a multi-operator multi-technology transport network within the European 5GEx Sandbox, including Segment Routing and optical domains. 

One of the main features of 5G networks is the coordinated orchestration of both information technology (IT) and connectivity resources. This enables the deployment of a flexible and programmable architecture able to provision end-to-end services with different (and sometimes quite) stringent quality of service (QoS) constraints. In this paradigm, service orchestration may take place over single or multiple administrative domains. The 5G Exchange (5GEx) project, building on the software-defined network and the network function virtualization (NFV) concepts, targets the design and the implementation of a multi-domain orchestrator (MdO) prototype for the automatic provisioning of network service across multiple administrative domains. This paper presents an architectural solution, designed and implemented in the context of the 5GEx MdO prototype, that can be used to establish end-to-end connectivity tunnels with QoS constraints (i.e., bandwidth and/or end-to-end delay) connecting virtualized network functions deployed in remote data centers controlled by different providers by means of an innovative stateful backward recursive path-computation-element based computation procedure. The proposed solution has been experimentally validated in terms of scalability, reliability, and end-to-end workflow proof-of-concept. Results show how the designed solution permits the automatic establishment of QoS-based end-to-end tunnels spanning across multi-technology and multi-operator network domains. The orchestration scheme does not present scalability issues for either the advertisement of resources or for the provisioning of connectivity services. Moreover, no issues have been identified from the reliability point of view.

Network slicing enables an infrastructure provider (InP) to support heterogeneous 5G services over a common platform (i.e., by creating a customized slice for each service). Once in operation, slices can be dynamically scaled up/down to match the variation of their service requirements. An InP generates revenue by accepting a slice request. If a slice cannot be scaled up when required, an InP has to also pay a penalty (proportional to the level of service degradation). It becomes then crucial for an InP to decide which slice requests should be accepted/rejected in order to increase its net profit. This paper presents a slice admission strategy based on reinforcement learning (RL) in the presence of services with different priorities. The use case considered is a 5G flexible radio access network (RAN), where slices of different mobile service providers are virtualized over the same RAN infrastructure. The proposed policy learns which are the services with the potential to bring high profit (i.e., high revenue with low degradation penalty), and hence should be accepted. The performance of the RL-based admission policy is compared against two deterministic heuristics. Results show that in the considered scenario, the proposed strategy outperforms the benchmark heuristics by at least 23%. Moreover, this paper shows how the policy is able to adapt to different conditions in terms of 1) slice degradation penalty versus slice revenue factors, and 2) proportion of high versus low priority services.

A failure in optical backbone network can cause tremendous consequences as a substantial number of connections often each carrying a large amount of data can be interrupted. Therefore, high reliability performance is essential for the network operators. Many existing works that aim at improving network reliability performance implicitly assume that the lifetime of devices is constant and independent of the traffic load. However, the reliability performance of a device is related to its occupancy. For example, the failure rate of erbium doped fiber amplifier (EDFA) can be expressed as a function of the number of amplified wavelengths. On the other hand, the choice of routing and wavelength assignment (RWA) algorithm impacts the link load and, as a consequence, can influence the number of EDFA failures in the network. In this paper we examine how RWA can impact the failure reparation related network operational costs. Several types of RWA approaches are considered, namely load-balancing, energy-awareness, and reliability-awareness. Among all the considered RWA algorithms, the reliability-aware RWA (RA-RWA) approach leverages on EDFA reliability profile to reduce the number of EDFA failures in the network and the related operational costs. The simulation results show that the RWA algorithm impacts in a significant way the operational costs caused by EDFA failures. The cost associated with reparation of an EDFA decreases by 7.8% (in case of RA-RWA) and increases by up to 40% (in case of a load-balancing approach) compared to the classical Shortest Path (SP) approach. Moreover, the cost caused by connection rerouting due to link unavailability triggered by EDFA failure exhibits a 20% decrease (RA-RWA) and up to 94% increase (energy-aware algorithm). We also analyze some key network performance metrics that may be affected by RWA, including blocking probability, link occupancy distribution, and path length.