Design of a Robust Search Algorithm for P2P Networks

Lecture Notes in Computer Science, 2004

In this paper we report a novel and efficient algorithm for searching p2p networks. The algorithm, termed ImmuneSearch, draws its basic inspiration from natural immune systems. It is implemented independently by each individual peer participating in the network and is totally decentralized in nature. ImmuneSearch avoids query message flooding; instead it uses an immune systems inspired concept of affinitygoverned proliferation and mutation for message movement. In addition, a protocol is formulated to change the neighborhoods of the peers based upon their proximity with the queried item. This results in topology evolution of the network whereby similar contents cluster together. The topology evolution coupled with proliferation and mutation help the p2p network to develop 'memory', as a result of which the search efficiency of the network improves as more and more individual peers perform search. Moreover, the algorithm is extremely robust and its performance is stable in face of the transient nature of the constituent peers.

Lecture Notes in Computer Science, 2005

Decentralized peer to peer (p2p) networks like Gnutella are attractive for certain applications because they require no centralized directories and no precise control over network topology or data placement. The greatest advantage is the robustness provided by them. However, flooding-based query algorithms used by the networks produce enormous amounts of traffic and substantially slow down the system. Recently, flooding has been replaced by more efficient k-random walkers and different variants of such algorithms. In this paper, we report immune-inspired algorithms for searching peer to peer networks. The algorithms use the immune-inspired mechanism of affinity-governed proliferation to spread query message packets in the network. Through a series of experiments, we compare the proliferation mechanism with different variants of random walk algorithms. The detailed experimental results show message packets undergoing proliferation spread much faster in the network and consequently proliferation algorithms produce better search output in p2p networks than random walk algorithms. Moreover, theoretical results by calculating the packet spreading speeds are reported which provide an understanding of the improved performance of the proliferation based search algorithm.

Proceedings of the eleventh international conference on Information and knowledge management - CIKM '02, 2002

One important problem in peer-to-peer (P2P) networks is searching and retrieving the correct information. However, existing searching mechanisms in pure peer-to-peer networks are inefficient due to the decentralized nature of such networks. We propose two mechanisms for information retrieval in pure peer-to-peer networks. The first, the modified Breadth-First-Search (BFS) mechanism, is an extension of the current Gnuttela protocol, allows searching with keywords, and is designed to minimize the number of messages that are needed to search the network. The second, the Intelligent Search mechanism, uses the past behavior of the P2P network to further improve the scalability of the search procedure. In this algorithm, each peer autonomously decides which of its peers are most likely to answer a given query. The algorithm is entirely distributed, and therefore scales well with the size of the network. We implemented our mechanisms as middleware platforms. To show the advantages of our mechanisms we present experimental results using the middleware implementation.

Sufficient uptime is the main concern in a distributed system where there is no centralized organization and control. Therefore searching in P2P file sharing systems is still a big problem due to the high churn rate of nodes. To find a particular piece of data within the network P2P systems explicitly or implicitly provide a lookup mechanism, which largely depends on the availability of nodes. In this paper we propose a stable node based ranking technique by detecting reliable nodes with adequate uptime by nature to ensure efficient searching and un-interruption during the data receiving. We have simulated an unstructured file-sharing environment to evaluate our proposed technique.

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.

In order to use Internet resources efficiently we need to search and locate information efficiently. System performance diminishes by either duplicating a large quantity of data on each and every node or flooding query to all the nodes in the network. Firstly, this paper reviews various searching algorithms. Search techniques can be classified as blind search in which information about neighbors is not kept by the peer and informed search where peers store information for routing queries to other nodes. It discusses how range queries can be processed efficiently by rotating scheme over structured P2P systems and secure searching algorithm based on topology adaptation which penalizes the malicious peers. Genetic algorithm providing parallel search are also covered in the paper. Lastly, it focuses on merits, demerits and applicability of these algorithms in different situations.

International Journal of Computer Applications, 2013

Peer-to-Peer (P2P) [1] are widely used for file sharing purposes. This type of usage provides decentralized solutions over centralized complex architecture. Peer-to-Peer networks are gaining attention from both the scientific perspective as well as the large Internet community. Popular applications utilizing this new technology offer many attractive features to a growing number of users. P2P is an architecture which is all-together a different class of applications that use the concept of distributed resources to perform an important crucial function in a decentralized manner. The popularity and bandwidth consumption attributed to current Peer-to-Peer filesharing applications makes the operation of these distributed systems very important for the Internet community. Efficiently discovering the queried resource is the initial and most important step in establishing an efficient peer-to-peer communication. Here, we will be describing and analyzing the performances of some existing search mechanisms deployed for the peer discovery and the content look up.

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.

2008

Abstract:,We consider network,clustering as the way to improve,the perfor- mance,of locating data in unstructured,P2P systems. Connectivity-based Dis- tributed node Clustering (CDC), and SCM-based Distributed Clustering (SDC) are two major,protocols that allow partitioning a network,topology,into clus- ters, based on node connectivity. These protocols focus on the accuracy of the clustering scheme, i.e. using the Scale Coverage Measure (SCM), and its maintenance

IEICE Transactions on Communications, 2006

The most prevalent peer-to-peer (P2P) application till today is file sharing, and unstructured P2P networks can support inherent heterogeneity of peers, are highly resilient to peers' failures, and incur low overhead at peer arrivals and departures. Dynamic querying (DQ) is a new flooding technique which could estimate a proper time-to-live (TTL) value for a query flooding by estimating the popularity of the searched files, and retrieve sufficient results under controlled flooding range for reducing network traffic. Recent researches show that a large amount of peers in the P2P file sharing system are the free-riders, and queries are seldom hit by those peers. The free-riding problem causes a large amount of redundant messages in the DQ-like search algorithm. In this paper, we proposed a new search algorithm, called "AntSearch," to solve the problem. In AntSearch, each peer maintains its hit rate of previous queries, and records a list of pheromone values of its immediate neighbors. Based on the pheromone values, a query is only flooded to those peers which are not likely to be the free-riders. Our simulation results show that, compared with DQ and its enhanced algorithm DQ+, the AntSearch algorithm averagely reduces 50% network traffic at almost the same search latency as DQ+, while retrieving sufficient results for a query with a given required number of results. key words: peer-to-peer network, dynamic querying, flooding, free-riding Kai-Hsiang Yang received a B.A. degree in