Christof Leng

Bubblestorm: resilient, probabilistic, and exhaustive peer-to-peer search

Peer-to-peer systems promise inexpensive scalability, adaptability, and robustness. Thus, they are an attractive platform for file sharing, distributed wikis, and search engines. These applications often store weakly structured data, requiring sophisticated search algorithms. To simplify the search problem, most scalable algorithms introduce structure to the network. However, churn or violent disruption may break this structure, compromising search guarantees. This paper proposes a simple probabilistic search system, BubbleStorm, built on random multigraphs. Our primary contribution is a flexible and reliable strategy for performing exhaustive search. BubbleStorm also exploits the heterogeneous bandwidth of peers. However, we sacrifice some of this bandwidth for high parallelism and low latency. The provided search guarantees are tunable, with success probability adjustable well into the realm of reliable systems. For validation, we simulate a network with one million low-end peers and show BubbleStorm handles up to 90% simultaneous peer departure and 50% simultaneous crash.

Modelling the Internet Delay Space Based on Geographical Locations

Existing approaches for modelling the Internet delay space predict end-to-end delays between two arbitrary hosts as static values. Further, they do not capture the characteristics caused by geographical constraints. Peer-to-peer (P2P) systems are, however, often very sensitive to the underlying delay characteristics of the Internet, since these characteristics directly influence system performance. This work proposes a model to predict lifelike delays between a given pair of end hosts. In addition to its low delay computation time, it has only linear memory costs which allows large scale P2P simulations to be performed. The model includes realistic delay jitter, subject to the geographical position of the sender and the receiver. Our analysis, using existing Internet measurement studies reveals that our approach seems to be an optimal tradeoff between a number of conflicting properties of existing approaches.

Piki - A Peer-to-Peer based Wiki Engine

In this paper, Wiki technology proved to be an appropriate tool for sharing global knowledge, as it provides an easy way to contribute and consume information. The way wikis are used, e.g. consuming and contributing information, naturally matches the peer-to-peer (p2p) paradigm, which inherently follows the structure of its distributed users. Compared to client-server solutions, it has significantly lower maintenance costs, no bottleneck or single point of failure, and resources are used and offered by all users. We propose a purely p2p based wiki engine (PIKI) with the focus on concurrent editing, version control, and decentralized full-text search (in a structured overlay). Further we elaborate tracking semantic information between linked articles through different versions, where a typed link itself holds meta-data about the linked files.

Maintaining replicas in unstructured P2P systems

Replication is widely used in unstructured peer-to-peer systems to improve search or achieve availability. We identify and solve a subclass of replication problems where each object is associated with a maintainer node, and its replicas should only be available as long as its maintainer is part of the network. Such requirement can be found in various applications, e.g., when objects are directory lists, service lists, or subscriptions of a publish/subscribe system. We provide maintainers with proven guarantees on the number of replicas, in spite of network churn and crash failures. We also tackle the related problems of changing the number of replicas, updating replicas, balancing storage load in a heterogeneous network, and eliminating replicas left by crashing maintainers. Our algorithm is based on probabilistic methods and is simple to implement. We show by simulation and formal proof that our algorithm is correct.

Towards Benchmarking of Structured Peer-to-Peer Overlays for Network Virtual Environments

Network virtual environments (NVE) are an evolving trend combining millions of users in an interactive community. A distributed NVE platform promises to lower the administration costs and to benefit from research done in the peer-to-peer (p2p) domain. In order to reuse existing mature p2p overlays for NVEs, a comparative evaluation has to be done in the same environment (e.g. resources of peers, peer behavior, churn, etc.), using appropriate test cases (scenarios) and observing relevant performance metrics. In this paper we present a benchmarking approach for p2p overlays in the context of NVEs. We define related quality attributes, scenarios, and metrics and use them to evaluate Chord and Kademlia as most popular p2p overlays and assess their suitability to NVE.

Channel-based Unidirectional Stream Protocol (CUSP)

This paper presents a novel transport protocol, CUSP, specifically designed with complex and dynamic network applications in mind. Peer-to-peer applications benefit in particular, as their requirements are met by neither UDP nor TCP. While other modern transports like SCTP or SST have also tried to combine the advantages of TCP and UDP, CUSP overcomes their technical and conceptual shortcomings. CUSP makes it possible to directly express application logic in the message flow. Modern applications need a mixture of request-response, request-multiple-response, publish-subscribe, and message-passing. All of these operations can be conveniently implemented using CUSPs unidirectional streams. We separate low-level packet management from streams into reusable channels. A channel connects two applications providing negotiation, congestion control, and cryptography. Developers operate on the stream level, sending messages as reliable and ordered byte-streams. Although they may share a common channel, a stall or loss in one stream does not block the others.

Peer-to-Peer Based Version Control

Many software projects are developed by globally distributed teams. The nature of the peer-to-peer paradigm fits to such an application scenario. However, existing tools to support developer require a central instance, like it is the case in version control, which is crucial for software development. This paper address the challenges version control faces in a purely peer-to-peer environment and presents a peer-to-peer based version control system we developed. We will evolve this basic approach, adding more features like change-set version control.

An online gaming testbed for peer-to-peer architectures

In this demo we present a testbed environment for Peer-to-Peer (P2P) game architectures. It is based on Planet PI4, an online multiplayer game whose gameplay provides a standard workload for a set of gaming-specific network interfaces. Its pluggable architecture allows for the evaluation and comparison of existing and new P2P networking approaches. Planet PI4 can run on a real network for prototypical evaluation as well as in a discrete-event simulator providing a reproducible environment.

Practical summation via gossip

Distributed summation is computed in asymptotically minimal rounds by Kempe et als Push-Sum algorithm. Unfortunately it has minor problems in practise, resolved here.

Modeling User Behavior in P2P Systems

The evaluation of peer-to-peer (P2P) systems is crucial for understanding their performance and therefore their feasibility in the real world. Different techniques, such as testbeds, analytical analysis, and simulations, can be used to evaluate system performance. For peer-to-peer systems, simulations are often the most reasonable approach, simply because P2P systems are both inherently large-scale and complex.