Stochastic analysis of file-swarming systems

File swarming (or file sharing) is one of the most important applications in P2P networks. In this paper, we propose a stochastic framework to analyze a file-swarming system under realistic setting: constraints in upload/download capacity, collaboration among peers and incentive for chunk exchange. We first extend the results in the coupon system [L. Massoulie, M. Vojnovic, Coupon replication systems, in: Proc. ACM SIGMETRICS, Banff, Alberta, Canada, 2005] by providing a tighter performance bound. Then we generalize the coupon system by considering peers with limited upload and download capacity. We illustrate the last-piece problem and show the effectiveness of using forward error-correction (FEC) code and/or multiple requests to improve the performance. Lastly, we propose a framework to analyze an incentive-based file-swarming system. The stochastic framework we propose can serve as a basis for other researchers to analyze and design more advanced features of file-swarming systems.     

Analysis and scheduling of practical network coding in OFDMA relay networks

Network coding has become a prominent approach to improve throughput of wireless networks. However, most of work in the literature concentrates mainly on 802.11-like random access networks. New technologies such as OFDMA (orthogonal frequency division multiple access), offer new opportunities for employing network coding. This paper considers how to apply the practical network coding scheme in OFDMA relay networks via cross-layer optimization. Specifically, we aim to explore the following questions: (1) When and how can wireless nodes select relay paths in the presence of network coding? (2) How can an OFDMA relay system assign network resource such as subcarrier and power for all the transmitting nodes? (3) What are the impacts of OFDMA system parameters on the network coding gain? To answer these questions, two efficient coding-aware relay strategies are presented to select forwarding paths with fixed and dynamic power allocation. In order to exploit the network capacity in slow frequency selective fading channels, we formulate optimization frameworks and propose channel-aware coding-aware resource allocation algorithms for an arbitrary traffic pattern. Our studies show that the network coding (i.e. XOR) gain depends on the nodes’ powers, traffic patterns etc. Especially, OFDMA relay network with dynamic power possesses both coding gain and power gain. Extensive simulations are performed to verify our analysis and demonstrate the throughput improvement of our proposals in the presence of XOR coding.     

On oligopoly spectrum allocation game in cognitive radio networks with capacity constraints

Dynamic spectrum sharing is a promising technology to improve spectrum utilization in future wireless networks. The flexible spectrum management provides new opportunities for licensed primary user and unlicensed secondary users to reallocate the spectrum resource efficiently. In this paper, we present an oligopoly pricing framework for dynamic spectrum allocation in which the primary users sell excessive spectrum to the secondary users for monetary return. We present two approaches, the strict constraints (type-I) and the QoS penalty (type-II), to model the realistic situation that the primary users have limited capacities. In the oligopoly model with strict constraints, we propose a low-complexity searching method to obtain the Nash Equilibrium and prove its uniqueness. When reduced to a duopoly game, we analytically show the interesting gaps in the leader–follower pricing strategy. In the QoS penalty based oligopoly model, a novel variable transformation method is developed to derive the unique Nash Equilibrium. When the market information is limited, we provide three myopically optimal algorithms “StrictBEST”, “StrictBR” and “QoSBEST” that enable price adjustment for duopoly primary users based on the Best Response Function (BRF) and the bounded rationality (BR) principles. Numerical results validate the effectiveness of our analysis and demonstrate the convergence of “StrictBEST” as well as “QoSBEST” to the Nash Equilibrium. For the “StrictBR” algorithm, we reveal the chaotic behaviors of dynamic price adaptation in response to the learning rates.     

The Design Trade-Offs of BitTorrent-Like File Sharing Protocols

The BitTorrent (BT) file sharing protocol is very popular due to its scalability property and the built-in incentive mechanism to reduce free-riding. However, in designing such P2P file sharing protocols, there is a fundamental trade-off between keeping fairness and providing good performance. In particular, the system can either keep peers (especially those resourceful ones) in the system for as long as possible so as to help the system to achieve better performance, or allow more resourceful peers to finish their download as quickly as possible so as to achieve fairness. The current BT protocol represents only one possible implementation in this whole design space. The objective of this paper is to characterize the design space of BT-like protocols. The rationale for considering fairness in the P2P file sharing context is to use it as a measure of willingness to provide service. We show that there is a wide range of design choices, ranging from optimizing the performance of file download time, to optimizing the overall fairness measure. More importantly, we show that there is a simple and easily implementable design knob so that the system can operate at a particular point in the design space. We also discuss different algorithms, ranging from centralized to distributed, in realizing the design knob. Performance evaluations are carried out, both via simulation and network measurement, to quantify the merits and properties of the BT-like file sharing protocols.     

An online framework for catching top spreaders and scanners

Flow level information is important for many applications in network measurement and analysis. In this work, we tackle the “Top Spreaders” and “Top Scanners” problems, where hosts that are spreading the largest numbers of flows, especially small flows, must be efficiently and accurately identified. The identification of these top users can be very helpful in network management, traffic engineering, application behavior analysis, and anomaly detection.We propose novel streaming algorithms and a “Filter-Tracker-Digester” framework to catch the top spreaders and scanners online. Our framework combines sampling and streaming algorithms, as well as deterministic and randomized algorithms, in such a way that they can effectively help each other to improve accuracy while reducing memory usage and processing time. To our knowledge, we are the first to tackle the “Top Scanners” problem in a streaming way. We address several challenges, namely: traffic scale, skewness, speed, memory usage, and result accuracy. The performance bounds of our algorithms are derived analytically, and are also evaluated by both real and synthetic traces, where we show our algorithm can achieve accuracy and speed of at least an order of magnitude higher than existing approaches.     

A Distributed Throttling Approach for Handling High Bandwidth Aggregates

Public-access networks need to handle persistent congestion and overload caused by high bandwidth aggregates that may occur during times of flooding-based DDoS attacks or flash crowds. The often unpredictable nature of these two activities can severely degrade server performance. Legitimate user requests also suffer considerably when traffic from many different sources aggregates inside the network and causes congestion. This paper studies a family of algorithms that "proactively" protect a server from overload by installing rate throttles in a set of upstream routers. Based on an optimal control setting, we propose algorithms that achieve throttling in a distributed and fair manner by taking important performance metrics into consideration, such as minimizing overall load variations. Using ns-2 simulations, we show that our proposed algorithms 1) are highly adaptive by avoiding unnecessary parameter configuration, 2) provide max-min fairness for any number of throttling routers, 3) respond very quickly to network changes, 4) are extremely robust against extrinsic factors beyond the system control, and 5) are stable under given delay bounds.     

Experiences of designing a sophisticated network monitor

This paper describes the design of a network monitor that captures connections information in a network. The real-time performance requirement makes the design significantly more challenging than that of a simple monitor that only counts packet types. Although we developed the monitor for DECnet, most of the lessons we learned are protocol-independent and are thus applicable to designing such a monitor for a different protocol. In addition to reviewing the functionality, we report a first-cut performance evaluation of the monitor software. We also describe some of our early experiences in using the monitor and review some current applications of the monitor in the areas of network configuration planning and performance management.     

Profiling and identification of P2P traffic

Accurate identification of network applications is important for many network activities. The traditional port-based technique has become much less effective since many new applications no longer use well-known fixed port numbers. In this paper, we propose a novel profile-based approach to identifying traffic flows belonging to the target application. In contrast to the method used in previous studies, of classifying traffic based on statistics of individual flows, we build behavioral profiles of the target application, which describe dominant patterns in the application. Based on the behavior profiles, a two-level matching method is used to identify new traffic. We first determine whether a host participates in the target application by comparing its behavior with the profiles. Subsequently, we compare each flow of the host with those patterns in the application profiles to determine which flows belong to this application. We demonstrate the effectiveness of our method on-campus traffic traces. Our results show that one can identify popular P2P applications with very high accuracy.     

On the interaction of multiple overlay routing

In the past few years, overlay networks have received much attention but there has been little study on the “interaction” of multiple, co-existing overlays on top of a physical network. In addition to previously introduced concept of overlay routing strategy such as selfish routing, we introduce a new strategy called “overlay optimal routing”. Under this routing policy, the overlay seeks to minimize its weighted average delay by splitting its traffic onto multiple paths. We establish that (i) the overlay optimal routing can achieve better delay compared to selfish routing and (ii) there exists a Nash equilibrium when multiple overlays adopt this strategy. Although an equilibrium point exists for overlay optimal routing and possibly for selfish routing, we show that the interaction of multiple overlay routing may not be Pareto optimal and that some fairness anomalies of resource allocation may occur. This is worthy of attention since overlay may not know the existence of other overlays and they will continue to operate at this sub-optimal point. We explore two pricing schemes to resolve the above issues. We show that by incorporating a proper pricing scheme, the overlay routing game can be led to the desired equilibrium and avoid the problems mentioned above. Extensive fluid-based simulations are performed to support the theoretical claims.     

Entropy Based Adaptive Flow Aggregation

Internet traffic flow measurement is vitally important for network management, accounting and performance studies. Cisco's NetFlow is a widely deployed flow measurement solution that uses a configurable static sampling rate to control processor and memory usage on the router and the amount of reporting flow records generated. But during flooding attacks the memory and network bandwidth consumed by flow records can increase beyond what is available. Currently available countermeasures have their own problems: 1) reject new flows when the cache is full - some legitimate new flows will not be counted; 2) export not-terminated flows to make room for new ones - this will exhaust the export bandwidth; and 3) adapt the sampling rate to traffic rate - this will reduce the overall accuracy of accounting, including legitimate flows. In this paper, we propose an entropy based adaptive flow aggregation algorithm. Relying on information-theoretic techniques, the algorithm efficiently identifies the clusters of attack flows in real time and aggregates those large number of short attack flows into a few metaflows. Compared to currently available solutions, our solution not only alleviates the problem in memory and export bandwidth, but also significantly improves the accuracy of legitimate flows. Finally, we evaluate our system using both synthetic trace file and real trace files from the Internet.