Technologies in software-defined networks (SDNs) introduce programmatic ways to reorganise the network logical topology. A possible practical usage of SDNs is reactive routing, where the logical topology is continuously evolving based on traffic statistics and policies. Usually, the SDNs controllers are considered transparent to the higher layers. It is expected that changes in logical topology may not affect applications. The goal is to study the impact of logical topology changes on BitTorrent, a popular peer-to-peer protocol in practice. This study focuses on BitTorrent, and the experimental results show that BitTorrent may produce the opposite effect to the one expected. The authors have run 32 BitTorrent clients in an emulated SDN ring topology and changed the virtual topology periodically by removing one link at the time from the ring. The experiments produced lower propagation when logical topology changed periodically than when it was static for BitTorrent traffic. For comparison, the same experiments were recreated using HTTP. For HTTP, slower propagation is obtained when logical topology changed than when it was static. Finally, the results are discussed and it has been concluded that high layer protocols need to be carefully studied, and in some cases adapted, before being deployed in SDNs.

Software defined networking (SDN) and network function virtualisation (NFV) have become hot topics in recent years. On one hand, SDN decouples the control plane from the data plane allowing the rapid innovation and the introduction of new services in an easy way. On the other hand, currently proprietary appliances such as load balancers and firewalls are implemented in hardware, NFV aims to change these network functions to an open software environment using virtualisation and cloud technologies. This means a reduction of spends in the provisioning and management of telecom services. SDN and NFV are two different concepts but these can coexist and help each other. In this study, the authors present a survey of SDN and NFV focusing in virtualisation projects and the use cases where a synergy between these technologies is possible. This study includes the basic concepts of network virtualisation, NFV and SDN, current research and the relation between both technologies.

Since the cloud paradigm becomes increasingly popular for dynamic resources allocation, the flexibility of a cloud is still limited regarding network services and their autonomous federation between different providers. The following architectural approach introduces a generic layered model to orchestrate and federate heterogeneous networks. In particular, an architecture is presented that enables quality of service (QoS) aware parameterisation of network resources in a cloud infrastructure of a single data-centre as well as for a federation. Furthermore, this architecture uses a service-level agreement (SLA) protocol and language to expose key performance indicators and to negotiate appropriate QoS constrains that are applied to the virtually sliced underlying network substrate. In this way, capabilities of the orchestration and the current utilisation of the network are building the foundation for dynamic negotiated SLAs and the within-guaranteed QoS network resources. Therefore an aggregation mechanism is illustrated for merging service-level objectives and for guaranteeing a single SLA that specifies obligations and responsibilities of all participants.

Cloud-based virtual networking environments are required to provide fine-grained quality of service (QoS) control without sacrificing scalability. However, no single approach can currently achieve these two goals simultaneously. FlowVisor is a building block to virtualise networks with fine-grained QoS support; however, scalability issues caused by the OpenFlow protocol and the centralised control model are a major concern. This study introduces a distributed FlowVisor (DFVisor) platform to address these scalability issues. The proposed DFVisor uses a layered overlay mechanism to improve network addressing space and switch capacity. DFVisor uses a distributed synchronised two-level database system with a synchronisation mechanism to enable the centralised control functions in the current FlowVisor platform in distributed control modules within the virtual network controllers. Therefore it removes a single point of failure in the network and reduces the flow setup latency without sacrificing the centralised network configuration and management capabilities. More importantly, the proposed DFVisor platform enables an advanced push-based flow setup and statistics collection mechanism to address scalability issues caused by the current pull-based flow setup and statistics collection method. A DFVisor prototype and an evaluation of this distributed synchronised two-level database are presented, and key issues for future research are discussed.

OpenFlow (OF) is one of the most widely used protocols for controller-to-switch communication in a software defined network (SDN). Performance analysis of OF-based SDN using analytical models is both highly desirable and challenging. There already exists a very elegant analytical model based on M/M/1 queues to estimate the packet sojourn time and probability of lost packets for the case in which a controller is responsible for only a single node in the data plane. However the literature falls short when it comes to the multiple node case, i.e. when there is more than one node in the data plane. In this work we propose a model to address this challenge by approximating the data plane as an open Jackson network with the controller also modeled as an M/M/1 queue. The model is then used to evaluate the system in the light of some of the metrics, such as; how much time a packet spends on average in an OF-based network and how much data we can pump into the network given the average delay requirements. Finally the PDF and the CDF of the time spent by the packet in an OF-based SDN for a given path is derived.