Sascha Grau

Towards the design of unexploitable construction mechanisms for multiple-tree based P2P streaming systems

In peer-to-peer based live streaming systems, a great number of participants have to cooperate to efficiently and reliably distribute a continuous flow of data. Each receiving peer in return provides its resources to the system. Since these systems operate in a completely distributed manner, it is of particular importance, to prevent malicious members from harvesting important topology information or influencing the streaming system to their needs. In this article, we analyze potential attack methods on multiple-tree-based P2P streaming systems, discuss important design decisions to constrain the impact of malicious behaviour, and we introduce the new concept of peer testaments. By analyzing existing systems, we show that so far only few attention has been given to the design of unexploitable construction mechanisms. Based on the identified design decisions, we propose a novel streaming system and evaluate it by exposing it to different types of internal attackers. Our results show that these attackers have to spend large effort to reach relevant positions in the streaming topology and that their bandwidth contribution far outnumbers the damage they achieve.

On Complexity and Approximability of Optimal DoS Attacks on Multiple-Tree P2P Streaming Topologies

We investigate the hardness of malicious attacks on multiple-tree topologies of push-based Peer-to-Peer streaming systems. In particular, we study the optimization problem of finding a minimum set of target nodes to achieve a certain damage objective. For this, we differentiate between three natural and increasingly complex damage types: global packet loss, service loss when using Multiple Description Coding, and service loss when using Forward Error Correction. We show that each of these attack problems is NP-hard, even for an idealized attacker with global knowledge about the topology. Despite tree-based topologies seem susceptible to such attacks, we can even prove that (under strong assumptions about NP) there is no polynomial time attacker, capable of guaranteeing a general solution quality within factors of c1 log(n) and c22log1-δn (with n topology nodes, δ = 1/log logd n for d <; 1/2 and constants c1, c2), respectively. To our knowledge, these are the first lower bounds on the quality of polynomial time attacks on P2P streaming topologies. The results naturally apply to major real-world DoS attackers and show hard limits for their possibilities. In addition, they demonstrate superior stability of Forward Error Correction systems compared to Multiple Description Coding and give theoretical foundation to properties of stable topologies.

Methods for Improving Resilience in Communication Networks and P2P Overlays

ABSTRACT Resilience to failures and deliberate attacks is becoming an essential requirement in most communication networks today. This also applies to P2P Overlays which on the one hand are created on top of communication infrastructures, and therefore are equally affected by failures of the underlying infrastructure, but which on the other hand introduce new possibilities like the creation of arbitrary links within the overlay. In this article, we present a survey of strategies to improve resilience in communication networks as well as in P2P overlay networks. Furthermore, our intention is to point out differences and similarities in the resilience-enhancing measures for both types of networks. By revising some basic concepts from graph theory, we show that many concepts for communication networks are based on well-known graph-theoretical problems. Especially, some methods for the construction of protection paths in advance of a failure are based on very hard problems, indeed many of them are in NP and can only be solved heuristically or on certain topologies. P2P overlay networks evidently benefit from resilience-enhancing strategies in the underlying communication infrastructure, but beyond that, their specific properties pose the need for more sophisticated mechanisms. The dynamic nature of peers requires to take some precautions, like estimating the reliability of peers, redundantly storing information, and provisioning a reliable routing.

Resilient and underlay-aware P2P live-streaming

Application Layer Multicast (ALM) represents a cost-efficient way to disseminate content in large scale. However, as it relies on end-systems in content distribution, it can be easily attacked and thus requires specific measures to increase its resilience against attacks. Besides attacks on end-users, few attention has been paid to attacks on the underlying transport network so far. When the overlay is not constructed in an underlay-aware manner, several overlay links may rely on the same link or router in the underlay. Hence, a single underlay failure may result in multiple, simultaneous overlay failures. Moreover, without considering the underlying transport network an inefficient content distribution can be the result. For this reason the ALM induced traffic load in transport networks can become rather large. In this article, we propose a construction algorithm for ALM topologies that incorporates information about the underlying network to improve their resilience against underlay failures, to maintain resilience against overlay attacks, and to increase the efficiency of the content distribution. Our simulation results indicate that the underlay dependence of the established ALM overlays can be nearly halved compared to overlays that do not use information about the transport network in their construction. As a result, the ALM induced traffic load in transport networks decreases considerably. In addition, the results indicate that our topologies are likewise resilient to underlay as well as overlay attacks.

On the Dependencies between Source Neighbors in Optimally DoS-stable P2P Streaming Topologies

We study tree-based peer-to-peer streaming topologies that minimize the maximum damage that can be caused by the failure of any number of peers. These optimally stable topologies can be characterized by a distinctive damage sequence. Although checking whether a given topology is optimally stable is a co-NP-complete problem, a large subclass of these topologies can be constructed by applying a simple set of rules. One of these rules states that every optimally stable topology must have optimally stable inter-dependencies between the nodes directly adjacent to the streaming source (called heads). However, until now, only a single stable head topology was known. In this article, we first give a short outline to previous results about optimally stable topologies. Then, we identify necessary and sufficient requirements for the optimal stability of head topologies, thereby largely increasing the number of known representatives from this class. All requirements can be checked in polynomial time. Furthermore, we show how to efficiently decide stability for head topologies with at most four stripes and give a procedure that, given a stable topology, produces a stable topology with an arbitrary number of stripes. Reversing this procedure can also speed up stability testing. Finally, we describe strategies how stable head topologies can be constructed in real-world streaming systems.

Attacker Independent Stability Guarantees for Peer-2-Peer-Live-Streaming Topologies

Live streaming systems based on ALM technology provide an effective and scalable way for content dissemination. Especially push-based systems, maintaining an explicit distribution topology, represent a promising approach. Compared to a classical client-server content distribution and in contrast to other ALM systems, they introduce only low additional transport delay and jitter. However, these systems are vulnerable towards random node failures or intended attacks. In this article, we select packet loss as a natural damage measure and show that there exists an attacker-independent concept of optimal stability of streaming topologies in push approaches. We are able to identify such optimally stable topologies by minimizing an upper bound on the sequence of damages any attacker may achieve on a topology. Furthermore, we show an easily constructible subclass of these topologies and present the unexpected result that deciding membership in the complete class of optimally stable topologies is a co-NP-complete problem.

Underlay-robust application layer multicast

In recent years, ALM emerged as cost-efficient and scalable form of content distribution by overcoming the classical client-server bottleneck. The client bandwidth is incorporated to stream distribution, so that every client that receives the stream forwards it as well. ALM systems are usually classified in push, pull and hybrid approaches [1]. In the remainder of this article we are focusing on live-streaming, which imposes strict delay constraints on the content distribution and thus cannot be realized with pull-based approaches. Hence, we concentrate on push-based ALM that splits a stream in multiple substreams (so-called stripes) by using Multiple Description Coding (MDC) and assigns each of them a separate spanning tree. However, the results of this article apply to hybrid and partially to pull-based approaches as well.

On virtualization-based network support for peer-assisted live-streaming applications

IPTV will be one of the main drivers of network innovation in the near future. Its bandwidth demands, especially if combined with HDTV, require smart algorithms for content distribution. One approach that is currently discussed intensively is Application Layer Multicast (ALM), which allows to scale to large node numbers by incorporating the resources of participating peers. In addition, locality-aware ALM approaches can distribute content close to the edges of the network and thus are able to relieve core networks from traffic. In this work, we propose to combine ALM with virtual routers that assist in content distribution, so-called ALM-Routers. They can be moved between distinct network positions and are inspired by current research towards a future Internet. Furthermore, we propose adaptation algorithms that operate solely on local knowledge and optimize ALM-Router positions as well as their overlay neighborhood. As our simulation results indicate, already a small fraction of such virtual nodes deployed on core network routers, decrease the overall traffic amount. Moreover, a subsequent optimization of the positions of ALM-Routers and the optimization of their overlay neighborhood induces additional considerable traffic savings compared to a static placement.

Dominating an s - t -Cut in a Network

We study an optimization problem with applications in design and analysis of resilient communication networks: given two vertices s, t in a graph G = (V,E), find a vertex set X ⊂ V of minimum cardinality, such that X and its neighborhood constitute an s-t vertex separator. Although the problem naturally combines notions of graph connectivity and domination, its computational properties significantly differ from these relatives.

Attack-Resilient Multitree Data Distribution Topologies

We consider a scenario of information broadcast where a source node distributes data in parallel over a fixed number of trees spanning over a large audience of nodes. The trees used for data dissemination are called distribution topology. Particular implementations of this scenario are peer-to-peer live streaming systems. Encoding data partially redundant, nodes are satisfied as long as they receive packets in at least a certain portion of trees. Otherwise, they are called isolated.