Showing papers presented at "International Conference on Peer-to-Peer Computing in 2004"

Percolation search in power law networks: making unstructured peer-to-peer networks scalable

Abstract: We introduce a scalable searching protocol for locating and retrieving content in random networks with power-law (PL) and heavy-tailed degree distributions. The proposed algorithm is capable of finding any content in the network with probability one in time O(logN), with a total traffic that provably scales sub-linearly with the network size, N. Unlike other proposed solutions, there is no need to assume that the network has multiple copies of contents; the protocol finds all contents reliably, even if every node in the network starts with a unique content. The scaling behavior of the size of the giant connected component of a random graph with heavy tailed degree distributions under bond percolation is at the heart of our results. The percolation search algorithm can be directly applied to make unstructured peer-to-peer (P2P) networks, such as Gnutella, Limewire and other file-sharing systems (which naturally display heavy-tailed degree distributions and scale-free network structures), scalable. For example, simulations of the protocol on the limewire crawl number 5 network, consisting of over 65,000 links and 10,000 nodes, shows that even for this snapshot network, the traffic can be reduced by a factor of at least 100, and yet achieve a hit-rate greater than 90%.

A robust protocol for building superpeer overlay topologies

Abstract: The concept of superpeer has been introduced to improve the performance of popular file-sharing applications. A superpeer is a node in a P2P network that operates as a server for a set of clients, and as an equal w.r.t. other superpeers. By exploiting heterogeneity, the superpeer paradigm allows P2P networks to run more efficiently, without compromising their decentralized nature. This paper describes SG-1, a novel generic mechanism for the construction and the maintenance of overlay topologies based on superpeers. SG-1 is based on the well-known gossip paradigm, with nodes exchanging information with randomly selected peers and re-arranging the topology according to the requirements of the particular P2P application. The resulting protocol is extremely efficient and robust, capable to deal with a continuous flow of nodes joining and leaving the system, as well as to repair a network where up to 100% of the existing super-peers have been removed.

Erasure code replication revisited

Abstract: Erasure coding is a technique for achieving high availability and reliability in storage and communication systems. We revisit the analysis of erasure code replication and point out some situations when whole-file replication is preferred. The switchover point (from preferring whole-file replication to erasure code replication) is studied, and characterized using asymptotic analysis. We also discuss the additional considerations in building erasure code replication systems.

P2P-based collaborative spam detection and filtering

Abstract: Spam is one of the major problems of today email systems. While many solutions have been proposed to automatically detect and filter spam, spammers are getting more and more technically sophisticated and aware of internal workings of anti-spam systems, finding ways to disguise their emails to get around the different controls that can be enforced. We propose a decentralized privacy-preserving approach to spam filtering. Our solution exploits robust digests to identify messages that are a slight variation of one another and a structured peer-to-peer architecture between mail servers to collaboratively share knowledge about spam.

From selfish nodes to cooperative networks - emergent link-based incentives in peer-to-peer networks

Abstract: For peer-to-peer (P2P) systems to operate effectively peers need to cooperate for the benefit of the network as a whole. Most existing P2P systems assume cooperation, relying on peers to perform tasks that are of no direct individual benefit. However, when large open systems are deployed such assumptions no longer hold because by adapting selfishly nodes may become "freeloaders" leaching resources from the network. We present initial results from simulations of an algorithm allowing nodes to adapt selfishly yet maintaining high levels of cooperation in both a Prisoners' dilemma and a flood-fill query scenario. The algorithm does not require centralized or third party reputation systems, the monitoring of neighbor behavior or the explicit programming of incentives and operates in highly dynamic and noisy networks. The algorithm appears to emerge its own incentive structure.

Adaptive peer-to-peer topologies

Abstract: We present a peer-level protocol for forming adaptive, self-organizing topologies for data-sharing P2P networks. This protocol is based on the idea that a peer should directly connect to those peers from which it is most likely to download satisfactory content. We show that the resulting topologies are more efficient than standard Gnutella topologies. Furthermore, we show that these adaptive topologies have the added benefits of increased resistance to certain types of attacks, intrinsic rewards for active peers and punishments for malicious peers and free riders.

PeerStore: better performance by relaxing in peer-to-peer backup

Abstract: Backup is cumbersome. To be effective, backups have to be made at regular intervals, forcing users to organize and store a growing collection of backup media. In this paper we propose a novel peer-to-peer backup system, PeerStore, that allows the user to store his backups on other people's computers instead. PeerStore is an adaptive, cost-effective system suitable for all types of networks ranging from LAN, WAN to large unstable networks like the Internet. The system consists of two layers: metadata layer and symmetric trading layer. Locating blocks and duplicate checking is accomplished by the metadata layer while the actual data distribution is done between pairs of peers after they have established a symmetric data trade. By decoupling the metadata management from data storage, the system offers a significant reduction of the maintenance cost and preserves fairness among peers. Results show that PeerStore has a reduced maintenance cost comparing to pStore. PeerStore also realizes fairness because of the symmetric nature of the trades.

PKI-based security for peer-to-peer information sharing

Abstract: Freeflow of information is the feature that has made peer-to-peer information sharing applications popular. However, this very feature holds back the acceptance of these applications by the corporate and scientific communities. In these communities it is important to provide confidentiality and integrity of communication and to enforce access control to shared resources. We present a collection of security mechanisms that can be used to satisfy these security requirements. Our solutions are based on established and proven security techniques and we utilize existing technologies when possible. As a proof of concept, we have developed an information sharing system, called scishare, which integrates a number of these security mechanisms to provide a secure environment for information sharing. This system will allow a broader set of user communities to benefit from peer-to-peer information sharing.

Constructing a balanced, (log(N)/1oglog(N))-diameter super-peer topology for scalable P2P systems

Abstract: Current peer-to-peer (P2P) file sharing applications are remarkably simple and robust, but their inefficiency can produce very high network loads. The use of super-peers has been proposed to improve the performance of unstructured P2P systems. These have the potential to approach the performance and scalability of structured systems, while retaining the benefits of unstructured P2P systems. There has, however, been little consensus on the best topology for connecting these super-peers, or how to construct the topology in a distributed, robust way. In this paper we propose a scalable unstructured P2P system (SUPS). The unique aspect of SUPS is a protocol for the distributed construction of a super-peer topology that has highly desirable performance characteristics. The protocol is inspired by the theory of random graphs. We describe the protocol, and demonstrate experimentally that it produces a balanced and low-diameter super-peer topology at low cost. We show that the method is very robust to super-peer failures and inconsistent information, and compare it with other approaches.

Reputation management framework and its use as currency in large-scale peer-to-peer networks

Abstract: We propose a reputation management framework for large-scale peer-to-peer (P2P) networks, wherein all nodes are assumed to behave selfishly. The proposed framework has several advantages. It enables a form of virtual currency, such that the reputation of nodes is a measure of their wealth. The framework is scalable and provides protection against attacks by malicious nodes. The above features are achieved by developing trusted communities of nodes whose members trust each other and cooperate to deal with the problem of nodes' selfishness and possible maliciousness.

Broose: a practical distributed hashtable based on the de-Bruijn topology

Abstract: Broose is a peer-to-peer protocol based on the de Bruijn topology allowing a distributed hashtable to be maintained in a loose manner. Each association is stored on k nodes to allow higher reliability with regard to node failures. Redundancy is also used when storing contacts avoiding complex topology maintenance for node departures and arrivals. It uses a constant size routing table of 0(k) contacts for allowing lookups in O(log N) message exchange (where N is the number of nodes participating). It can also be parameterized for obtaining O(log N / log log N) steps lookups with a routing table of size 0(k log N). These bounds hold with high probability. Moreover, the protocol allows load balancing of hotspots of requests for a given key as well as hotspots of key collisions. The goal is to obtain a protocol as practical as Kademlia based on the de Bruijn topology.

Collaborative automated trust negotiation in peer-to-peer systems

Abstract: The increasing popularity of peer-to-peer systems has promoted the development of new techniques to support various kinds of business transactions. However, users are reluctant to conduct high value transactions over P2P systems due to the inherent untrustworthiness of peers. In this paper, we investigate building trust through automated trust negotiation, which is orthogonal to existing reputation-based approaches. Trust negotiation makes it possible to prove a peer satisfies certain trust requirements imposed by the ongoing business. We also introduce locally trusted third parties, which we demonstrate can effectively promote successful trust negotiations in peer-to-peer systems.

Aggregating information in peer-to-peer systems for improved join and leave

Abstract: We introduce the Distributed Approximative System Information Service (DASIS) as a useful scheme to aggregate approximative information on the state of a peer-to-peer system. We present how this service can be integrated into existing peer-to-peer systems, such as Kademlia and Chord. As a sample application, we show how DASIS can be employed for establishing an effective deterministic join algorithm. Through simulation, we demonstrate that the insertion of peers using DASIS information results in a well-balanced system. Moreover, our join algorithm gracefully resolves load imbalances in the system due to unfortunate biased leaves of peers.

A distributed and measurement-based framework against free riding in peer-to-peer networks

Abstract: We propose a distributed and measurement-based method to reduce the degree of free riding in P2P networks. We primarily focus on developing schemes to locate free riders and on determining policies that can be used to take actions against them. Our proposed schemes require each peer to monitor its neighboring peers, make decisions if they exhibit any kind of free riding, and take appropriate actions if required.

ICN: interest-based clustering network

Abstract: An interest-based clustering peer-to-peer network (ICN) architecture is introduced in this paper. ICN uses a lot of Freenet mechanisms and is based on cache management. ICN is self-organizing, fully distributed, scalable, and logically hierarchical. In ICN, the upper level is bound by de Bruijn graph. Nodes in the lower level self-cluster based on interest. Through analysis and simulation, ICN shows good fault-tolerance, efficient data retrieval and resource usage as well as low overhead traffic.

SeAl: managing accesses and data in peer-to-peer sharing networks

Abstract: We present SeAl, a novel data/resource and data-access management infrastructure designed for the purpose of addressing a key problem in P2P data sharing networks, namely the problem of wide-scale selfish peer behavior. Selfish behavior has been manifested and well documented and it is widely accepted that unless this is dealt with, the scalability, efficiency, and the usefulness of P2P sharing networks will be diminished. SeAl essentially consists of a monitoring/accounting subsystem, an auditing/verification subsystem, and incentive mechanisms. The monitoring subsystem facilitates the classification of peers into selfish/altruistic. The auditing/verification layer provides a shield against perjurer/slandering and colluding peers that may try to cheat the monitoring subsystem. The incentives mechanisms effectively utilize these layers so to increase the computational/networking and data resources that are available to the community. Our extensive performance results show that SeAl performs its tasks swiftly, while the overhead introduced by our accounting and auditing mechanisms in terms of response time, network, and storage overheads are very small.

Reliable peer-to-peer end system multicasting through replication

Abstract: A key challenge in peer-to-peer computing system is to provide decentralized and yet reliable services on top of a network of loosely coupled, weakly connected and possibly unreliable peers. This work presents an effective dynamic passive replication scheme designed to provide reliable multicast service in peer-cast, an efficient and self-configurable peer-to-peer end system multicast (ESM) system. We first describe the design of a distributed replication scheme, which enables reliable subscription and multicast dissemination of information in an environment of inherently unreliable peers. Then we present an analytical model to discuss its fault tolerance properties, and report a set of initial experiments, showing the feasibility and the effectiveness of the proposed approach.

On de Bruijn routing in distributed hash tables: there and back again

Abstract: We show in this paper that de Bruijn networks, despite providing efficient search while using constant routing table size, as well as simplicity of the understanding and implementation of such networks, are unsuitable where key distribution will be uneven, a realistic scenario for most practical applications. In presence of arbitrarily skewed data distribution, it has only recently been shown that some traditional P2P overlay networks with non-constant (typically logarithmic) instead of constant routing table size can meet conflicting objectives of storage load balancing as well as search efficiency. So this paper, while showing that de Bruijn networks fail to meet these dual objectives, opens up a more general problem for the research community as to whether P2P systems with constant routing table can at all achieve the conflicting objectives of retaining search efficiency as well as storage load balancing, while preserving key ordering (which leads to uneven key distribution).

Highways: proximity clustering for scalable peer-to-peer network

Abstract: The "location-aware" construction of overlay networks requires the identification of nodes that are efficient with respect to network delay and available bandwidth. In this short paper, we propose Highways to create clusters of nodes using a novel "location-aware" method, based on a scalable and distributed network coordinate system. This helps to build overlay routing tables to achieve better proximity accuracy, thus, providing a mechanism to boost performance in application overlay routing.

Information leak in the Chord lookup protocol

Abstract: In peer-to-peer (P2P) systems, it is often essential that connected systems (nodes) relay messages which did not originate locally, on to the greater network. As a result, an intermediate node might be able to determine a large amount of information about the system, such as the querying tendencies of other nodes. This represents an inherent security issue in P2P networks. Therefore, we ask the following question: through the observation of the network traffic in a P2P network, what kind of information can an adversarial node learn about another node in the same network? We study this question in the case of a specific P2P system - Chord. We also study the effects of the parameters of Chord (such as finger-table size) and the various enhancements to Chord (such as location caching and data caching) on the amount of information leaked.

A framework for testing distributed systems

Abstract: Thorough testing of distributed systems, particularly peer-to-peer systems can prove difficult due to the problems inherent in deploying, controlling and monitoring many nodes simultaneously. This problem will only increase as the scale of distributed systems continues to grow. This framework implements a test bed environment using a semi-centralized peer-to-peer network as a substrate for sharing resources made available from standard PCs. This framework automates the process of test-case deployment using a combination of reflection and aspect oriented programming. This allows 'point-and-click' publishing of software onto the test-bed. Our framework also provides a common monitoring, control and logging interface for all nodes running on the network. Together, these features greatly reduce deployment-time for real-world test scenarios. Automated insertion and removal of test code also ensures that the testing process does not compromise the correctness of the final system.

Update propagation through replica chain in decentralized and unstructured P2P systems

Abstract: We propose a novel algorithm, called update propagation through replica chain (UPTReC), to maintain file consistency in decentralized and unstructured peer-to-peer (P2P) systems. In UPTReC, each file has a logical replica chain composed of all replica peers (RPs) which are defined as peers that have replicas of the file. Each RP acquires partial knowledge of the bi-directional chain by keeping a list of information about k nearest RPs, called probe peers, in each direction. When an RP initiates an update, it pushes the update to all possible online (active) RPs through the replica chain. A reconnected RP pulls an online RP to synchronize the replica status and the information of the probe peers. An analytical model is derived to evaluate the performance of the UPTReC algorithm. The analytical results provide a better understanding of the system in choosing the system parameters for probabilistically guaranteed file consistency with minimum overheads. Simulation experiments are conducted to compare the performance with an existing update propagation algorithm based on the rumor spreading scheme. The experimental results show that the UPTReC can significantly reduce (up to 70%) overhead messages and also achieve smaller stale query ratio for files prone to frequent updates.

Cache updates in a peer-to-peer network of mobile agents

Abstract: In open multi-agent systems, agents need resources provided by other agents but they are not aware of which agents provide particular resources. We consider a peer-to-peer approach, in which each agent maintains a local cache with information about k resources, that is for each of the k resources, an agent that provides it. However, when an agent or a resource moves, cache entries become obsolete. We propose a suite of cache update policies that combine pull-based invalidation that is initiated by the agent that maintains the cache with push-based invalidation that is initiated by the agent that moves. We study and compare variations of oblivious flooding-based push/pull along with an informed push approach where each agent maintains a list of the agents that have it cached. Our experimental results indicate that a novel variation of flooding for push where a moving agent propagates its new location to agents in its old neighborhood achieves good cache consistency with a small message overhead. The proposed policies are suitable for any peer-to-peer system where peers cache information about other peers and this information becomes obsolete.

PROST: a programmable structured peer-to-peer overlay network

Abstract: We present the idea of a programmable structured P2P architecture. Our proposed system allows the key-based routing infrastructure, which is common to all structured P2P overlays, to be shared by multiple applications. Furthermore, our architecture allows the dynamic and on-demand deployment of new applications and services on top of the shared routing layer.

Semantic driven hashing (SDH): an ontology-based search scheme for the semantic aware network (SA Net)

Abstract: The success of personalized resource discovery depends on its ability to allow users to discover, extract and integrate information of interest from heterogeneous sources, and its ability to provide these users with efficient tools to manipulate and convert the discovered information into knowledge. We propose semantic aware network (SA Net), a structured peer-to-peer (P2P) overlay architecture to support the basic functionalities of a personalized resource discovery. The SA Net uses an ontology-based representation of the resources to enable a semantic resource discovery and access that reflects the interest of the user. We describe the SA Net search scheme "semantic-driven hashing " (SDH), which uses lexical-based ontology to provide a foundation for indexing and search in structured P2P overlay infrastructure.

The different dimensions of dynamicity

Abstract: In a peer-to-peer (P2P) network, a 'fabric' of overlay links helps peers discover and use other peers' resources. This fabric, however, is highly dynamic and constantly changing. While different measures of dynamicity have been implicitly and explicitly proposed, there is lack of deeper understanding about the various aspects of dynamicity. We systematically evaluate and quantify different dimensions of dynamicity through controlled generation of different dynamic networks. We also introduce a new dimension of dynamicity, persistence, which quantifies stable nodes in a network. This measure could be quite useful in the design and testing of P2P protocols that exploit the presence of stable nodes. Also, quite coincidentally, and to our surprise, a certain type of dynamic network that we designed has node degree properties that resemble those observed in social networks.

A placement scheme for peer-to-peer networks based on principles from geometry

Abstract: Crucial for the performance of peer-to-peer networks based on geometric topologies is the measurement complexity and quality of the mapping function used to map a node in the network to a point in the geometric target space. In this paper we study how results from mathematics as well as data mining can be applied to this mapping problem. Using a metric space model for networks and results from mathematics a relation between the number of nodes to be mapped, the worst case error of the mapping and the dimension of the geometric target space is formulated. As a main result geometric cluster placement (GCP) is presented, an improved and resilient placement algorithm based on GNP. An evaluation of GCP presented is based on measurement data from the RIPE NCC test traffic measurement (TTM) project.

Semantic-laden peer-to-peer service directory

Abstract: The most intuitive way to build a service directory application that allows for service entities to register or search for services on top of a structured peer-to-peer network is to build reverse indices at appropriate nodes on the network. However, this implies trust on the reliability and integrity of other nodes on the network, which may be too risky an assumption for businesses. This paper proposes a service directory that groups service entities of the same category together; this is achieved by dedicating part of the node identifiers to correspond to their service category semantic. Using chord as the peer-to-peer substrate, this scheme logically divides the chord circle into equidistant arcs; each arc is called an island. This scheme results in the formation of islands of varying population, and thus changing the uniformly spread topology of the original chord. Simulations are used to investigate the path length and message load of the changed topology. An additional routing scheme is also proposed and simulated to exploit the new topology to gain better path length.

A framework for developing reflective and dynamic P2P networks (RaDP2P)

Abstract: There has been a significant body of research conducted into various structured and unstructured overlay network protocols. Both paradigms have advantages for specific application domains and researchers are beginning to examine the benefits of using hybridized systems. We hypothesize that resource awareness and adaptation are essential to the efficient exploitation of the resources available on the diverse nodes which compose peer-to-peer networks. To support this, we propose a hybrid peer-to-peer model which uses an unstructured decentralised network layered on top of a structured overlay to provide support for multiple levels of adaptation.

P2P-4-DL: digital library over peer-to-peer

Abstract: The P2P-4-DL project aims to investigate and build a DL system that would operate over a P2P structure Rather than storing digital objects centrally they remain the responsibility of the individual peers that provide them This allows the system to utilise network resources more efficiently as well as providing users with a greater sense of control over the digital objects they share Our prototype also draws upon natural language processing (NLP) techniques in an attempt to increase the usability of the system Other related work within this area includes EDUTELLA, a RDF based P2P infrastructure that can support the development of DL's