Marco Hoffmann

Applying NFV and SDN to LTE mobile core gateways, the functions placement problem

With the rapid growth of user data, service innovation, and the persistent necessity to reduce costs, todays mobile operators are faced with severe challenges. In networking, two new concepts have emerged aiming at cost reduction, increase of network scalability and service flexibility, namely Network Functions Virtualization (NFV) and Software Defined Networking (SDN). NFV proposes to run the mobile network functions as software instances on commodity servers or datacenters (DC), while SDN supports a decomposition of the mobile network into control-plane and data-plane functions. Whereas these new concepts are considered as very promising drivers to design cost efficient mobile network architectures, limited attention has been drawn to the network load and infringed data-plane delay imposed by introducing NFV or SDN. We argue that within a widely-spanned mobile network, there is in fact a high potential to combine both concepts. Taking load and delay into account, there will be areas of the mobile network rather benefiting from an NFV deployment with all functions virtualized, while for other areas, an SDN deployment with functions decomposition is more advantageous. We refer to this problem as the functions placement problem. We propose a model that resolves the functions placement and aims at minimizing the transport network load overhead against several parameters such as data-plane delay, number of potential datacenters and SDN control overhead. We illustrate our proposed concept along with a concrete use case example.

A Virtual SDN-Enabled LTE EPC Architecture: A Case Study for S-/P-Gateways Functions

The recent initiative of Network Functions Virtualization (NFV) aims to deliver any data- plane processing or control-plane function in high volume data centers or network elements to decrease operational cost and increase deployment flexibility. In order to dynamically direct traffic flows between respective network elements, Software Defined Networking (SDN) can be seen as one enabler. In this paper, we focus on mobile core network nodes such as the MME, HSS, S- and P- Gateway as standardized for the LTE Evolved Packet Core (EPC). One straightforward solution for a virtualized EPC architecture would be to move all EPC network nodes completely into a data center and handle the data traffic via SDN-enabled switches. However, this solution would keep the conventional monolithic architecture unchanged. A possible split in the EPC functionality between a centralized data center and operators transport network elements could be needed to provide the desired flexibility, performance and TCO reduction. Therefore, we have analyzed the EPC nodes and classified their functions according to their impact on data-plane and control-plane processing. We propose a mapping for these functions on four alternative deployment frameworks based on SDN and OpenFlow (OF). In addition, we investigate the current OF implementations capability to realize basic core operations such as QoS, data classification, tunneling and charging. Our analysis shows that functions, which involve high data packet processing such as tunneling, have more potential to be kept on the data-plane network element, i.e. realized by an OpenFlow Switch. We argue for an enhanced OF network element NE+, which contains additional network functions next to the basic OpenFlow protocol.

Heuristic Approaches to the Controller Placement Problem in Large Scale SDN Networks

Software Defined Networking (SDN) marks a paradigm shift towards an externalized and logically centralized network control plane. A particularly important task in SDN architectures is that of controller placement, i.e., the positioning of a limited number of resources within a network to meet various requirements. These requirements range from latency constraints to failure tolerance and load balancing. In most scenarios, at least some of these objectives are competing, thus no single best placement is available and decision makers need to find a balanced trade-off. This work presents POCO, a framework for Pareto-based Optimal COntroller placement that provides operators with Pareto optimal placements with respect to different performance metrics. In its default configuration, POCO performs an exhaustive evaluation of all possible placements. While this is practically feasible for small and medium sized networks, realistic time and resource constraints call for an alternative in the context of large scale networks or dynamic networks whose properties change over time. For these scenarios, the POCO toolset is extended by a heuristic approach that is less accurate, but yields faster computation times. An evaluation of this heuristic is performed on a collection of real world network topologies from the Internet Topology Zoo. Utilizing a measure for quantifying the error introduced by the heuristic approach allows an analysis of the resulting trade-off between time and accuracy. Additionally, the proposed methods can be extended to solve similar virtual functions placement problems which appear in the context of Network Functions Virtualization (NFV).

On routing and transmission-range determination of multi-bit-rate signals over mixed-line-rate WDM optical networks for carrier ethernet

Ethernets success in local area networks (LANs) is fueling the efforts to extend its reach to cover metro and long-haul networks. This new Ethernet is refereed to as Carrier Ethernet. Among the various transport infrastructures for realizing Carrier Ethernet, wavelength-division multiplexing (WDM) optical network is a strong candidate for this purpose. Optical transmission rates per channel are increasing from 10 to 40 Gb/s and even 100 Gb/s, and they can also coexist in the same fiber. Along with the flexibility associated with such a network with mixed-line rates (MLR), signal-related constraints at high rates become a challenge for cost-efficient routing. Among these issues is the maximum nonregenerated optical distance that a signal can travel before its quality degrades or maximum transmission range (TR). TR is rate-dependent: The higher the rate, the shorter the range. While high-rate pipes may require signal regeneration to restore the signals quality, they support more traffic and, hence, can save resources. We study the problem of cost-efficient routing of multi-bit-rate (1/10/40/100 Gb/s) Ethernet tunnels using MLR over a carriers WDM optical network with signal-transmission-range constraints. We studied the effect of TR for mixed-rate signals (10/40/100 Gb/s) on the networks cost to determine the optimal TR of each bit rate. We present an analytical model based on a mixed-integer linear program (MILP) to determine the optimal TR of a small network. Since MILP has scalability constraints that makes it hard or sometimes impossible to solve for real network topologies, we propose a graph-based solution that constructs a mixed-line-rate auxiliary (MLR-AUX) graph to capture the networks heterogeneity and a weight-assignment approach that allows the routing to be cost-efficient. Our algorithms were tested on a U.S. nationwide network topology. We found that it is possible to reduce the networks cost by using short TR and that the optimal TR depends strongly on traffic characteristics and on the TR values of different bit-rate signals.

A performance study of network migration to SDN-enabled Traffic Engineering

In this paper, we analyze the question of network migration to Software Defined Networking (SDN) from the perspective of Traffic Engineering (TE). For a given network topology and migration planning horizon, we ask the question of which routers in the IP network should migrate to SDN-enabled operation to reduce the need for network capacity upgrades over a given time horizon. We propose an algorithm to determine the optimum schedule for node substitution within a network, which shows that already a few SDN routers, when strategically located, provide a stably large number of path alternatives to be used in TE, thus substantially reducing the need for large network capacity upgrades.

Provisioning and Operation of Virtual Networks

In todays Internet, requirements of services regarding the underlying transport network are very diverse. In the future, this diversity will increase and make it harder to accommodate all services in a single network. A possible approach to keep up with this diversity in future networks is the deployment of isolated, custom tailored networks on top of a single shared physical substrate. The COMCON (COntrol and Monitoring of COexisting Networks) project aims to define a reference architecture for setup, control, and monitoring of virtual networks on a provider- and operator-grade level. In this paper, we present the building blocks and interfaces of our architecture.

SDN Partitioning: A Centralized Control Plane for Distributed Routing Protocols

Hybrid IP networks that use both control paradigms - distributed and centralized - promise the best of two worlds: 1) programmability of software-defined networking (SDN) and 2) reliability and fault tolerance of distributed routing protocols, like open shortest path first (OSPF). Typically, a hybrid network deploys SDN to control prioritized traffic and OSPF to assure care-free operation of best effort traffic. We propose a new hybrid network architecture, called SDN partitioning (SDNp), which establishes centralized control over the distributed routing protocol by partitioning the topology into sub-domains with SDN-enabled border nodes. In our approach, OSPFs routing updates have to traverse SDN border nodes to reach neighboring sub-domains. This allows the central controller to modify how sub-domains view one another, which in turn allows to steer inter-sub-domain traffic. The degree of dynamic control against simplicity of OSPF can be traded off by adjusting the size of the sub-domains. This paper studies the technical requirements, presents a novel scheme for balanced topology partitioning, and provides the corresponding network management models for SDNp. Our performance evaluation shows that - already in its minimum configuration with two sub-domains - SDNp provides significant improvements in all respects compared to legacy routing protocols, whereas smaller sub-domains provide network control capabilities comparable to full SDN deployment.

Demonstrating the optimal placement of virtualized cellular network functions in case of large crowd events

This demonstration shows how Network Function Virtualisation (NFV) [1] can be used by a network provider to dynamically provide required mobile core network functions in case of a large ”Mega” event like a soccer game or a music festival. Economical reasons may not justify the installation or continuous maintenance of expensive dedicated hardware which is necessary to cope with the high load generated by visitors of such an event only in some parts of the network and only for a short time. The Evolved Packet Core (EPC) in nowadays’ mobile LTE networks consists of several, specialized components: first pure control-plane elements like Mobility Management Entities (MME) which can be installed on virtualized IT hardware in the cloud already today and second gateways which are a mixture of controland user-plane. This demonstration focuses on the Serving Gateway (SGW) which switches GTP (GPRS Tunneling Protocol) tunnels in a LTE network. In such an LTE network with ≈ 10 mio. subscribers about 10 SGW devices are in use. On the way to a successful deployment of NFV-based EPC components, several challenges have to be met. This includes the deployment, interconnection and configuration of LTE components in the cloud. As such, entities that instantiate and orchestrate the virtualized functions are required. The presented demonstration shows a scenario of NFVbased dynamic capacity addition to a LTE mobile network, indicated in Figure 1. By incorporating the new demands added by the increased access capacity, virtualized SGW instances are launched at the optimal locations in the network, by re-programming SDN enabled network elements (NE+). Figure 1(a) shows the normal configuration of the LTE network including the statically located SGW devices as well as

An adaptive inter-domain PCE framework to improve resource utilization and reduce inter-domain signaling

Upcoming broadband commercial and scientific applications are now demanding high bandwidth pipes across multiple domains with guaranteed Quality of Service (QoS). Recent research initiatives such as the Path Computation Element (PCE) framework are focusing on the development of scalable multi-domain QoS provisioning frameworks, especially within the emerging carrier grade transport technologies based on layer-2 tunnels. QoS provisioning across multiple domains requires that QoS parameters for available transit paths inside a domain be advertised in the inter-domain routing algorithms, while the dynamic inter- and intra-domain connections vary the available resource, and hence require frequent inter-domain updates. The signaling load on the other hand hampers the scalability of the inter-domain routing mechanisms. We propose the use of an adaptive partitioning framework, which can effectively use network resources and at the same time stabilize the advertised domain topologies and thus path advertisements. Our method partitions network resources by pre-reserving resources for inter-domain transit traffic, and uses policies to modify the resource partitioning in order to maintain the available transit capacity between specified bounds. We show by simulations that the proposed mechanism can reduce inter-domain signaling load by 10%-20% and reduce overall blocking inside a domain by creating a trade-off between available resources for intra-domain connections and inter-domain transit connections. The reduction in inter-domain signaling and blocking can be used as a building block to design scalable QoS routing systems for carrier grade transport networks.

Towards a Cost Optimal Design for a 5G Mobile Core Network Based on SDN and NFV

With the rapid growth of user traffic, service innovation, and the persistent necessity to reduce costs, today’s mobile operators are faced with several challenges. In networking, two concepts have emerged aiming at cost reduction, increase of network scalability and deployment flexibility, namely Network Functions Virtualization (NFV) and Software Defined Networking (SDN). NFV mitigates the dependency on hardware, where mobile network functions are deployed as software virtual network functions on commodity servers at cloud infrastructure, i.e., data centers. SDN provides a programmable and flexible network control by decoupling the mobile network functions into control plane and data plane functions. The design of the next generation mobile network (5G) requires new planning and dimensioning models to achieve a cost optimal design that supports a wide range of traffic demands. We propose three optimization models that aim at minimizing the network load cost as well as data center resources cost by finding the optimal placement of the data centers as well the SDN and NFV mobile network functions. The optimization solutions demonstrate the trade-offs between the different data center deployments, i.e., centralized or distributed, and the different cost factors, i.e., optimal network load cost or data center resources cost. We propose a Pareto optimal multi-objective model that achieves a balance between network and data center cost. Additionally, we use prior inference, based on the solutions of the single objectives, to pre-select data center locations for the multi-objective model that results in reducing the optimization complexity and achieves savings in run time while keeping a minimal optimality gap.