Hybrid Search Schemes for Unstructured Peer-to-Peer Networks

… Computing Systems, 2007. ICDCS'07. 27th …, 2007

In unstructured peer-to-peer (P2P) networks, the overlay topology (or connectivity graph) among peers is a crucial component in addition to the peer/data organization and search. Topological characteristics have profound impact on the efficiency of search on such unstructured P2P networks as well as other networks. It has been well-known that search on small-world topologies of N nodes can be as efficient as O(ln N ), while scale-free (power-law) topologies offer even better search efficiencies like as good as O(ln ln N ) for a range of degree distribution exponents. However, generation and maintenance of such scale-free topologies are hard to realize in a distributed and potentially uncooperative environments as in the P2P networks.

2005

Performing efficient decentralized search is a fundamental problem in Peer-to-Peer (P2P) systems. There has been a significant amount of research recently on developing robust self-organizing P2P topologies that support efficient search. In this paper we discuss four structured and unstructured P2P models (CAN, Chord, PRU, and Hypergrid) and three characteristic search algorithms (BFS, k-Random Walk, and GAPS) for unstructured networks.

Pdpta, 2006

Scalability in a peer-to-peer network is a challenging problem. Unstructured peer-to-peer networks inherently lack scalability, and structured networks are inefficient for a high churn rate. In this paper, we present a scalable search algorithm for a decentralized unstructured peer-to-peer network using a method to dynamically determine the number of nodes to forward a query to at once. The decision is based on the degree to which each neighbor has contributed to previous successful searches. The algorithm automatically creates a spanning graph of the high traffic links. Once a stable spanning graph is created, a query tends to travel along the edges of the spanning graph. This way, the number of hops required for a search is roughly bound by the diameter of the spanning graph. The simulation shows that our algorithm demonstrates significantly better performance in terms of the number of messages generated and hops required for a search over other popular algorithms.

2002

Decentralized and unstructured peer-to-peer networks such as Gnutella are attractive for certain applications because they require no centralized directories and no precise control over network topology or data placement. However, the flooding-based query algorithm used in Gnutella does not scale; each query generates a large amount of traffic and large systems quickly become overwhelmed by the queryinduced load. This paper explores, through simulation, various alternatives to Gnutella's query algorithm, data replication strategy, and network topology. We propose a query algorithm based on multiple random walks that resolves queries almost as quickly as Gnutella's flooding method while reducing the network traffic by two orders of magnitude in many cases. We also present simulation results on a distributed replication strategy proposed in . Finally, we find that among the various network topologies we consider, uniform random graphs yield the best performance. jaded observers the explosive increase in Peer-to-Peer (P2P) network usage has been astounding. Within a few months of Napster's introduction in 1999 the system had spread widely, and recent measurement data suggests that P2P applications are having a very significant and rapidly growing impact on Internet traffic . It is important to study the performance and scalability of these P2P networks.

Concurrency and Computation: Practice and Experience, 2008

This paper develops a directed generalized random graphs based analytical modeling framework to compare several variations of the basic flooding search strategy in unstructured decentralized peer-topeer networks. To validate the model predictions, we designed and implemented a distributed crawler architecture that is able to efficiently capture snapshots of the top-level Gnutella overlay topology. The snapshots are used to obtain simulation results that are used to assess the accuracy of our model. The model predictions are then used to compute system-oriented performance indexes (the average and the coefficient of variation of the number of query messages) as well as user-oriented measures (the probability of finding at least one replica of a resource, the average search time). The trade-off between the optimization of system-oriented measures and the improvement of user-oriented quality indexes is investigated for several variations of the basic flooding strategy suggesting that adding control parameters to the basic flooding mechanism might prove beneficial in this class of systems.

Cluster Computing, 2013

Peer-to-peer systems are important Internet applications. A major portion of Internet traffic belongs to such applications. Flooding search is a basic search scheme for unstructured peerto-peer networks, where a node must send a query message to all its neighbors when seeking a file (in a file sharing situation). Flooding has no knowledge about network topology and files distribution, thus it offers an attractive method for file discovery in dynamic and evolving networks. Although pure flooding can achieve high coverage but it produces exponentially redundant messages in each hop. Consequently, the growth of redundant messages limits system scalability and causes unnecessary traffic in networks. Besides, flooding has no opportunity to get an advantage of node diversity of participating in unstructured P2P networks. To improve this searching scheme and reduce redundant messages, this paper proposes a novel algorithm named HybridFlood. This algorithm is divided into two steps. The first step follows the flooding with a limited number of hops. In the second step, nosey nodes are selected in each searching horizon. The nosey nodes are nodes which have the most links to other nodes. These nodes maintain the data index of all client nodes. We provided analytical studies for flooding and HybridFlood. The analytical results provided the best threshold point of hop for optimum coverage growth rate and redundant messages in flooding. It also proved in HybridFlood broadcasting messages are cut down at least an order of magnitude. Thus, the proposed algorithm extends the search efficiency by reducing redundant messages in each hop. The simulation experiments validated analytical results.

2004

Searching for objects in unstructured peerlo-peer (P2P) networks is an important problem, and one that has received recent attention. In this paper, we present a simple, elegant, yet highly effective technique for object location (including rare objects). Our scheme installs object references at a known numhu of randomly selected peers. A query to the object is routed to a prede-Icnnined number of random peers, selected independently of (he installation procedure. The high probability of a non-empty intersection between these two sets forms the basis for our search mechanism. We prove analyticnUy, and demonstrate experimentally, that our scheme provides high probabilistic guarantees of success, while incurring minimal overhead. EITedive realiz.ation of the approach builds on a number of recent results on generating random walks, and efficiently estimating network size for unstructured networks. The presence of failures (departures) in the network pose additional challenges. Finally, effective strategies for installing references (0 replicas are critical for optimizing the lradeoffbetween installation overhead and search lime. We address these issues in an analytical framework and validate our results on a variety of real and synthetic topologies. Our results generalize to related problems of estimating and controlling object replication, and eliminaling duplicates in large-scale unstructured networks.

Parallel and Distributed Systems, IEEE …, 2009

In unstructured peer-to-peer (P2P) networks, the overlay topology (or connectivity graph) among peers is a crucial component in addition to the peer/data organization and search. Topological characteristics have profound impact on the efficiency of a search on such unstructured P2P networks, as well as other networks. A key limitation of scale-free (power-law) topologies is the high load (i.e., high degree) on a very few number of hub nodes. In a typical unstructured P2P network, peers are not willing to maintain high degrees/loads as they may not want to store a large number of entries for construction of the overlay topology. Therefore, to achieve fairness and practicality among all peers, hard cutoffs on the number of entries are imposed by the individual peers, which limits scale-freeness of the overall topology, hence limited scale-free networks. Thus, it is expected that the efficiency of the flooding search reduces as the size of the hard cutoff does. We investigate the construction of scale-free topologies with hard cutoffs (i.e., there are not any major hubs) and the effect of these hard cutoffs on the search efficiency. Interestingly, we observe that the efficiency of normalized flooding and random walk search algorithms increases as the hard cutoff decreases.

In this poster we will present our work on the design of efficient and reliable unstructured peer-to-peer (P2P) systems. Our work focuses on creating well-connected unstructured P2P overlay that can perform efficient searching and message routing. We show that well designed systems can tolerate high node failures (>25%) while maintaining connectivity and still resolving searches with few messages. This is important in applications such as file sharing and content distribution where there are many thousands of participating nodes that are widely dispersed and network coniditions are highly variable. Structured P2P systems are not suitable for these applications since such applications require multi-attribute and wild card searching. We show that carefully constructed overlays can resolve this type of search within 4 hops for large networks (> 10,000 nodes) with low object replication ratios (< 1%).

GLOBECOM '05. IEEE Global Telecommunications Conference, 2005., 2005

Search time as a function of the number of replicas of a queried object provides a key component to understanding system behavior in peer-to-peer networks. The analytical work in this area so far has assumed a uniform distribution of file replicas throughout the network with an implicit or explicit assumption of uniform file popularity distribution whereas, in reality, there is clear evidence of clustering in file popularity patterns. In this paper, we provide mechanisms for modeling clustering in file popularity distributions and the consequent non-uniform distribution of file replicas. We provide results for the search time in such networks for both random walk and flooding search mechanisms.