In pursuit of a flexible, resource efficient and high- performant 5G infrastructure, many operators, vendors and research consortia are currently developing, testing and inte­grating their NFV platform with associated management and orchestration (MANO) functionality. The SONATA NFV platform follows a micro-service design, which involves a tight coupling between an SDK, monitoring and MANO functionality, targeting a secure and stable software foundation. This experience paper gives a thorough overview on the encountered challenges, insights and resulting learnings when implementing and integrating the SONATA Service Platform using a continuous integration and delivery DevOps methodology. This is the result of a strong cooperation between prominent equipment vendors, network operators, software companies and universities, providing a set of constructive recommendations in hope of catalysing the development and deployment of NFV platforms.

Deploying applications in federated clouds is becoming increasingly important. The ability to place application components in various locations is appealing to a large set of distributed applications. Components spread across a number of clouds still need to communicate with each other. Existing solutions for interconnecting clouds are using only single paths between sites, even though clouds typically have multiple up-links. This paper introduces our work on a MultiPath TCP (MPTCP) proxy suited for cloud interconnection. The proxy transforms TCP streams into multipath connections using MPTCP. By virtue of MPTCP, the bandwidth of multiple links is made available to the proxied connections and resilience can be improved without the need for additional fail-over mechanisms. The proposed solution works transparently for applications, making changes to existing applications unnecessary. Measurements of the implemented scheme confirm the additional throughput achievable and the gain in resilience.

5G does not only aim at higher capacity and lower latency than current 4G in mobile broadband networks, but also to increase the level of programmability, control and flexibility to meet the requirements from innovative use cases such as IoT, smart manufacturing, and immersive media. However, several difficulties still need to be overcome for a better technological adoption. To reduce the time-to-market for networked services and to lower the entry barrier to third party developers of VNFs and network services, an integrated development and operations (DevOps) methodology is a promising way. One of the biggest challenges in upcoming 5G DevOps is the validation and verification (V&V) of individual VNFs and network services (VNF graphs) so that operators can be sure of their behavior. In this paper, we propose an NFV architecture that supports the DevOps methodology with a V&V platform and an advanced NFV catalogue. The conceptual components of the V&V platform, the foreseen methodology and the outcomes are explained. Finally, we explore some perspectives in applying such V&V approach in another 5G research project.

Cloud federation is receiving increasing attention due to the benefits of resilience and locality it brings to cloud providers and users. Our analysis of three diverse use cases shows that existing solutions are not addressing the federation needs of such use case applications. In this paper, we present an alternative approach to network federation, providing a model based on cloud-to-cloud agreements. In our scenarios, companies hosting their own OpenStack clouds need to run machines transparently in another cloud, provided by a company they have an agreement with. Our solution provides multiple benefits to cloud providers and users detailed in this paper. Our implementation outperforms the VPNaaS solution in OpenStack in terms of throughput.

Services are increasingly provided by cloud computing systems. Such services are often accessed while on the move. If the distance between the user and the service is getting large, quality-of-experience, in particular for interactive services, deteriorates. To counter this, the provisioning location of the service should be as close to the user as possible. This paper describes the concept of follow-me cloud, according to which services are migrating in unison with the user's movements. The key components required for such integrated user mobility/service mobility management are introduced and mappings to an OpenFlow-based implementation are described.

Non-functional properties are an essential constituent of service level agreements as they describe those quality-of-service parameters that are not related to the actual function of a service. Thus, non-functional properties let providers create distinguishing service offers and let consumers discriminate between various offers that provide the same function. The negotiation of non-functional properties is how service level agreements are commonly established. This chapter introduces various forms, models, specifications, and realizations of service level agreement negotiation to provide a broad background of the current state-of-the-art. Although different in various details, the described systems share a number of common features. Based on them, a holistic architecture is defined combining previous work into one coherent framework. The architecture is applicable to different negotiation models and protocols, and covers all functions of the negotiation phase. Based on this architecture, particular challenges and areas of future work are motivated. These mostly revolve around increasing the acceptance of service level agreement negotiation and enhancing interoperability.

Software Defined Networking (SDN in short) is reshaping the future of computer networks. By decoupling control and data planes, SDN technologies allow a more flexible management of network infrastructures, whose resources may be operated by means of a well defined programming interface. Several approaches have been recently proposed to implement the SDN concept. OpenFlow is maybe the most prominent SDN component, having been supported by several device vendors. This paper discusses a practical experience in designing an OpenFlowcontroller for a Mobile Cloud Management system. We present the programming model and the designed abstraction and discuss the lesson learned.

The management of the entire service landscape comprising a Cloud environment is a complex and challenging venture. There, one task of utmost importance, is the generation and processing of information about the state, health, and performance of the various services and IT components, something which is generally referred to as monitoring. Such information is the foundation for proper assessment and management of the whole Cloud. This chapter pursues two objectives: first, to provide an overview of monitoring in Cloud environments and, second, to propose a solution for interoperable and vendor-independent Cloud monitoring. Along the way, the authors motivate the necessity of monitoring at the different levels of Cloud infrastructures, introduce selected state-of-the-art, and extract requirements for Cloud monitoring. Based on these requirements, the following sections depict a Cloud monitoring solution and describe current developments towards interoperable, open, and extensible Cloud monitoring frameworks.

Ubiquitous network access allows people to access an ever increasing range of services from a variety of mobile terminals, including laptops, tablets and smartphones. A flexible and economically efficient way of provisioning such services is through Cloud Computing. Assuming that several cloud-enabled datacenters are made available at the edges of the Internet, service providers may take advantage of them by optimally locating service instances as close as possible to their users. By localizing traffic at the edges of access networks, such an approach may result beneficial for both service and network providers. In this paper we present Follow-Me Cloud (FMC), a technology developed at NEC Laboratories Europe that allows transparent migration of services in TCP/IP networks, thanks to the dynamic configuration of a set of coordinated OpenFlow switches located at the edge of the network. In particular, in this paper we analyze the scalability properties of an FMC-based system and propose a role separation strategy based on distribution of control plane functions which enables scale-out of the system. By means of simulation, we prove that the application of the proposed separation strategy results in less state retained by individual OpenFlow controllers and in more effective localization of network traffic.

Designing data centers with adequate performance and at the right scale is a difficult task. The load in terms of network and CPU utilization is hard to estimate without appropriate tools. In this paper we introduce a data center performance estimation system which can support planning data centers and application deployments. The tool takes data center topologies and applications and calculates the load expected on the server and networking infrastructure. To quickly produce results, the data center evaluator uses linear programming.