Locating and Storing Information in Peer-to-Peer Network

Abstract

In a peer-to-peer network configuration of systems, storing and locating files is a perplexing problem. Researchers have studied and implemented various techniques. All those proved to be useful, but they had their shortcomings and drawbacks. Lack of scalability and a lot of time consumption were the most crucial issues with those studies. In a few studies, the accuracy of results obtained also was an issue. In this paper, we study various data structures and algorithms for data storage and location in peer-to-peer systems. We study techniques for establishing efficient Routing Indices and evaluating their performance. Routing Indices is a useful way of locating files in a p2p system. This paper discusses three types of Routing Indices. These routing indices are compound, hop count and exponential. In this mechanism, nodes forward queries to their neighboring nodes. These nodes have their local-indexes to locate the information. We have defined techniques for creating, updating and maintaining these routing indices. We also study the occurrence of cycles in the network and ways to detect and recover from it.

Locating and Storing Information in Peer-to-Peer Network

Introduction

Network configurations are of two types namely, Peer-to-peer and client/server configuration. In a peer-to-peer configuration, each computer system connects to every other system by a direct link between them. On the contrary, in client/server configurations, there is no direct connection between each system in a network. Each system connects to a centralized system called server. All the computers in the network connect to the server. In this way, the computers get connected to each other. The P2P configuration of systems is emerging as the most useful configuration worldwide for association over the internet. This configuration of systems consists of a huge database of files to be exchanged between the systems. Peer store enormous amounts of files on the systems that can be accessed by other peers. Researchers have formulized various techniques for such issues. Document location and recovery are points of concern for the administrators of p2p networks.

Various problems exist for locating and storing information in a p2p network. Users face problems like inaccurate search results, enormous time-consumption by the network to achieve the results etc. In a peer-to-peer environment, people need to search their files or perform any operation. For this, they need to get to the correct node or peer of the network. The typical applications for data processing in a p2p network configuration can be sharing of files between peers. One of the peers may want to know the status of another peer. A peer may want to provide lecture to another peer. All these applications require the network to locate the address of the specified peer.

These issues need efficient resolution. Researchers have done various researches on this issue. In his paper, Donald Kossman has discussed distributed data processing. The author has discussed various techniques for query processing. The author suggested a few useful techniques. In spite of that, the paper failed to address the issue of scalability (Kossman, ...