In unstructured peer-to-peer networks, the average response latency and traffic cost of a query are two main performance metrics. Controlled flood resource query algorithms are widely used in unstructured networks, such as peer-to-peer networks. In this work, we propose a new algorithm called Selective Dynamic Query (SDQ). Based on mathematical programming, SDQ calculates the optimal combination of an integer TTL value and a set of neighbors to control the scope of the next query. Our results demonstrate that SDQ provides finer grain control than other algorithms: its response latency is close to the minimum known through Ring Expansion; Meanwhile, its traffic cost is also close to the minimum. To our best knowledge, this is the first job to achieve a better balance between the response latency and the cost of traffic.
Dynamic query (DQ) is a technique adopted in unstructured Peer-to-Peer (P2P) networks to minimize the number of peers that need to be visited to achieve the desired number of results. In this paper we present the use of the DQ technique in structured P2P networks. In particular, we present a P2P search algorithm, called DQ-DHT (dynamic query on a distributed hash table), to perform DQ-type searches on DHT-based overlays. The objective of DQ-DHT is twofold: to allow arbitrary queries on structured P2P networks and to provide a dynamic adaptation of the search according to the popularity of the resources to be located. This article describes the DQ-DHT algorithm using Chord as a basic overlay and analyzes its performance compared to DQ in unstructured networks.
The emergence of Napster, a P2P file-sharing system, at the end of 1998 is considered the true beginning of P2P systems. It attracted many people to share their music files and download music files from their peers. But now P2P is not limited to just sharing files. Instead, it refers to a class of systems and applications that use distributed resources to perform a critical function in a decentralized manner. The resources of many users and computers can be assembled to generate large groups of information and significant computing power. In a peer-to-peer (P2P) system, equal role nodes or capabilities exchange information and services directly with each other.
Each node can serve as a server and as a client it is therefore called a servent (server + client). In addition, because computers communicate directly with their peers, the bandwidth of the network is best used. However, there are often disadvantages inherent in P2P precisely because of its decentralized nature. For example, in Gnutella, users look for files flooding the network with queries and causing each computer to match their local disk. Clearly, this type of solution may have difficulty scaling for a large number of sites or complex queries. At Napster, on the other hand, users can not search files around the world; They simply search on a single server that has only indexed a fraction of the available files.