Andreas Kirstädter

Traffic models for future backbone networks – a service-oriented approach

In this paper we present a novel approach to assess the impact of new and existing services on traffic volume in current and future backbone networks. Several proposals to model traffic load in access and backbone networks exist in the literature. These proposals consider current Internet traffic like http, smtp, ftp and Peer-to-Peer (P2P). We expect, however, that there will be a change in traffic load for future networks caused by services like IP Television (IPTV), Video on Demand (VoD) and Virtual Private Networks (VPN). Additionally, population-based models may no longer be applicable due to the widespread use of service-providers and hierarchical routing through network peering points. Therefore, it is important to reassess future traffic volumes and traffic patterns and to identify those services that have the most impact on the networks. We model todays traffic volume of each of the described services and estimate future traffic volumes taking peering points into account. To illustrate the different traffic flows and to characterise the traffic distribution we apply our results to a Germany reference network. Copyright

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.

Path protection with explicit availability constraints for virtual network embedding

Network reliability and survivability are important features for hosting high-demanding services in virtual network environments. Since multiple virtual networks can share the physical resources of the underlying substrate, even a single failure in the substrate can affect a large number of virtual networks and the services they offer. To overcome the physical link failure impact on virtual network operations, backup path mechanisms are investigated in many related works for path protection. However, this one to one path protection cannot handle explicit path availability requirements. In this work, we present a virtual network embedding algorithm which provides path protection with explicit virtual link availability constraints. Evaluation results show that our algorithm performs close to theoretical thresholds and consumes less link resources when delivering a virtual network embedding solution in polynomial time compared to a conventional path backup algorithm.

On the trade-off between cost and availability of virtual networks

To minimize cost, Virtual Network Operators (VNOs) need to consider the required network availability already at the network design stage. One generic approach to reach the availability target is to select only high-quality physical network elements that offer high availability and consequently demand high expenses per element. The other generic approach to achieve high availability is to add protection capacity on the level of the virtual network based on lower cost components. In this paper, we analyze both alternatives with a simulation tool to answer the fundamental question how quality can be traded against capacity. For this purpose, we consider different network topologies and the influence of different parameters and provide a framework to find an optimal strategy between Mean Time Between Failures (MTBF) targets for the physical infrastructure and the usage of additional backup paths on the virtual network level.

Channel modelling for free-space optical inter-HAP links using adaptive ARQ transmission

Free-space optical (FSO) communication systems have seen significant developments in recent years due to growing need for very high data rates and tap-proof communication. The operation of an FSO link is suited to diverse variety of applications such as satellites, High Altitude Platforms (HAPs), Unmanned Aerial Vehicles (UAVs), aircrafts, ground stations and other areas involving both civil and military situations. FSO communication systems face challenges due to different effects of the atmospheric channel. FSO channel primarily suffers from scintillation effects due to Index of Refraction Turbulence (IRT). In addition, acquisition and pointing becomes more difficult because of the high directivity of the transmitted beam: Miss-pointing of the transmitted beam and tracking errors at the receiver generate additional fading of the optical signal. High Altitude Platforms (HAPs) are quasi-stationary vehicles operating in the stratosphere. The slowly varying but precisely determined time-of-flight of the Inter-HAP channel adds to its characteristics. To propose a suitable ARQ scheme, proper theoretical understanding of the optical atmospheric propagation and modeling of a specific scenario FSO channel is required. In this paper, a bi-directional symmetrical Inter-HAP link has been selected and modeled. The Inter-HAP channel model is then investigated via simulations in terms of optical scintillation induced by IRT and in presence of pointing error. The performance characteristic of the model is then quantified in terms of fading statistics from which the Packet Error Probability (PEP) is calculated. Based on the PEP characteristics, we propose suitable ARQ schemes.

Evaluation of data-center architectures for virtualized Network Functions

Network Functions Virtualization (NFV) is a recent industrial trend which gains a lot of attention from telecom operators and vendors. The NFV concept delivers network services using commercial off-the-shelf hardware and IT virtualization technologies, thus virtualizing entire classes of network node functions. Traditional and recent data-center architectures are mostly built and optimized for data storage or web based applications. However, NFV type applications have different characteristics with respect to network load and needed computing capacity, which may impact the data-center design. Therefore in this paper, we are motivated to model the characteristics of NFV type applications and investigate the suitability of current data-center architectures for such new applications.