Marcell Perényi

Identification and Analysis of Peer-to-Peer Traffic

Recent measurement studies report that a significant portion of Internet traffic is unknown. It is very likely that the majority of the unidentified traffic originates from peer-to-peer (P2P) applications. However, traditional techniques to identify P2P traffic seem to fail since these applications usually disguise their existence by using arbitrary ports. In addition to the identification of actual P2P traffic, the characteristics of that type of traffic are also scarcely known. The main purpose of this paper is twofold. First, we propose a novel identification method to reveal P2P traffic from traffic aggregation. Our method does not rely on packet payload so we avoid the difficulties arising from legal, privacy-related, financial and technical obstacles. Instead, our method is based on a set of heuristics derived from the robust properties of P2P traffic. We demonstrate our method with current traffic data obtained from one of the largest Internet providers in Hungary. We also show the high accuracy of the proposed algorithm by means of a validation study. Second, several results of a comprehensive traffic analysis study are reported in the paper. We show the daily behavior of P2P users compared to the non-P2P users. We present our important finding about the almost constant ratio of the P2P and total number of users. Flow sizes and holding times are also analyzed and results of a heavy-tail analysis are described. Finally, we discuss the popularity distribution properties of P2P applications. Our results show that the unique properties of P2P application traffic seem to fade away during aggregation and characteristics of the traffic will be similar to that of other non-P2P traffic aggregation.

On the identification and analysis of Skype traffic

Skype applies strong encryption to provide secure communication inside the whole Skype network. It also uses several techniques to conceal the traffic and the protocol. As a consequence, traditional port-based or payload-based identification of Skype traffic cannot be applied. In this paper, after an overview of the Skype P2P system, network entities and operation, we introduce novel algorithms to detect several types of communications (including voice calls primarily) that the Skype client initiates toward dedicated servers of the Skype network and other peers. The common point in these algorithms is that all of them are based on packet headers only and the extracted flow level information. We do not need information from packet payloads. The identification methods allow us to discover logged on Skype users and their voice calls. The whole identification process is scripted in Transact-SQL; it can thus be executed automatically on a prerecorded (offline) data set. We present identification results, analysis and comparison of data sets captured in mobile and fixed networks. We also present the validation of the algorithms in both network types. Copyright

Skype Traffic Identification

Skype uses strong encryption to secure communications inside the whole Skype network. Clients choose communication ports randomly. Therefore traditional port based or payload based identification of Skype traffic is not feasible. In this paper we introduce a novel flow dynamics based identification method to discover Skype host and voice calls as well. Our method uses only packet headers and the extracted flow level information. The whole identification process is scripted in Transact-SQL, thus it can be executed automatically. We also present the validation of the algorithm together with some analysis of the identification results.

Grooming-enhanced multicast in multilayer networks

In this paper we investigate and evaluate the performance of multicast routing in grooming capable multi-layer optical wavelength division multiplexing (WDM) networks. New wavelength-graph models are proposed for network equipments capable of optical-layer branching of light-paths. We evaluate the cost-effectiveness of electronic and optical-layer multicast and unicast as well. The high scalability of optical multicast against unicast is also showed. All routing and technical constraints are formulated in ILP and realized in our versatile simulator.

On the identification and analysis of p2p traffic aggregation

The main purpose of this paper is twofold. First, we propose a novel identification method to reveal P2P traffic from traffic aggregation. Our method is based on a set of heuristics derived from the robust properties of P2P traffic. We show the high accuracy of the proposed algorithm based on a validation study. Second, several results of a comprehensive traffic analysis, focusing on the differences between P2P and non-P2P traffic, are reported in the paper. Our results show that the unique properties of P2P application traffic seem to fade away during aggregation and characteristics of the traffic will be similar to that of other non-P2P traffic aggregation.

Regular reconfiguration of light-trees in multilayer optical networks

In this paper we study dynamically changing multicast trees (light-trees) in double-layer optical networks. In this scenario, the continuous changing of multicast endpoints causes the degradation of the tree. Therefore a huge amount of network resources can be spared by regular reconfiguration. The benefit of reconfiguration is investigated for different routing algorithms and reconfiguration periods.

Grooming node placement in switched multilayer networks

In switched multilayer optical networks the bandwidth required by a demand is usually less than the capacity of a �� -channel. Thus it is worth multiplexing more demands into a single �� -channel. This process is called traffic grooming. However, it is too expensive and not always necessary to build this functionality into all network nodes. At low network load, having fewer grooming-nodes it is sufficient to attain relatively low blocking probability. In this paper we investigate how to choose a given number of grooming capable nodes in the network to minimize blocking probability (blocking ratio). We introduce a model of an optical network and define the procedure of grooming traffic demands together. We describe three algorithms that choose which nodes to be grooming- capable. The first one is a greedy algorithm the second one is based on topological properties of nodes while the third searches for a minimal covering vertex set in a certain sense. The performance of these algorithms is then investigated. We also study how the number of grooming-enabled nodes influences the fairness of the network, i.e. the equality of provisioning demands with different bandwidth requirements.

Traffic grooming and power level tuning for physical impairment constrained routing

In both, metropolitan optical networks (MON) and long haul optical networks (LHON) the signal quality is often influenced by the physical impairments, therefore proper impairment constrained routing decisions are needed. In this paper we propose two new approaches that jointly perform, on the one hand, routing and wavelength assignment (RWA) and, on the other hand, either tuning the signal power of certain wavelength division multiplexed (WDM) channels or grooming the traffic of some WDM channels in nodes that are grooming capable. We evaluate the proposed optimization methods for both, single and multilayer networks. In the first case we assume that no signal regeneration is allowed along the path, only re-amplification, while in the more complex two-layer case we assume grooming in the electronic layer that implicitly performs 3R signal regeneration and wavelength conversion as well.

Joint grooming capability and wavelength number optimization

In switched optical networks traffic grooming is a key factor in efficient bandwidth utilization. Without it a single traffic demand would occupy a full wavelength channel. However, it is not necessary to install grooming capability in all nodes of a network. Furthermore it is also unnecessary to equip all grooming nodes with full grooming capability. We can reduce the costs by deploying only the necessary grooming capacity. From a design point of view this means that not only the grooming capable nodes should be selected, but also the necessary grooming port number is to be determined. In our paper we propose algorithms which use iterative simulations to solve this problem.

Network Resilience Requirements and Algorithms for Multicasting and Broadcasting Digital TV

In this paper we give an overview of the network architecture and of the resilience requirements for both, metro and core networks.