P2P Techniques for Decentralized Applications (Synthesis Lectures on Data Management)

P2P Techniques for Decentralized Applications (Synthesis Lectures on Data Management)

Esther Pacitti, Reza Akbarinia, Manal El-Dick

Language: English

Pages: 106

ISBN: 1608458229

Format: PDF / Kindle (mobi) / ePub


As an alternative to traditional client-server systems, Peer-to-Peer (P2P) systems provide major advantages in terms of scalability, autonomy and dynamic behavior of peers, and decentralization of control. Thus, they are well suited for large-scale data sharing in distributed environments. Most of the existing P2P approaches for data sharing rely on either structured networks (e.g., DHTs) for efficient indexing, or unstructured networks for ease of deployment, or some combination. However, these approaches have some limitations, such as lack of freedom for data placement in DHTs, and high latency and high network traffic in unstructured networks. To address these limitations, gossip protocols which are easy to deploy and scale well, can be exploited. In this book, we will give an overview of these different P2P techniques and architectures, discuss their trade-offs, and illustrate their use for decentralizing several large-scale data sharing applications.

Contents:
P2P Overlays, Query Routing, and Gossiping / Content Distribution in P2P Systems / Recommendation Systems / Top-k Query Processing in P2P Systems

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2 1.1.3 Super-peer . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3 1.1.4 Comparing P2P Overlays . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3 Query Routing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4 1.2.1 Query Routing in Unstructured Overlays .

or rate yet. Items with high similarity are recommended to the user. Notice that here the user profile is defined based on the semantics of the item the user has seen or rated. To measure the similarity between items, each item is identified by a set of features and attributes that are usually extracted from its content or description. Therefore, content-based filtering 3.1. OVERVIEW OF RECOMMENDATION 45 systems are designed mostly to recommend text-based items, or items that have text

that consists of keywords or topics. Peer categories of interests are either extracted implicitly from peer documents, or stated explicitly by the user. Whenever a peer submits a query, shortcuts are selectively exploited. According to their results, the shortcut links reduces response time as well as the number of exchanged messages. Similarly, as Sripanidkulchai et al. [2003a] and Upadrashta et al. [2005] propose, each peer p adds shortcut links to the peers that have most recently and

helped to pioneer the concept of participation in Web 2.0. Popular examples include the online encyclopedia Wikipedia that enables individuals to create and edit content (articles), social networking sites like Facebook, photo and video sharing sites like YouTube and Flickr, as well as wikis and blogs. Social networking is even allowing scientific groups to expand their knowledge base and share their theories which might otherwise become isolated and irrelevant. With the Internet reaching a

1. P2P OVERLAYS, QUERY ROUTING, AND GOSSIPING Figure 1.6: Butterfly routing geometry. Hybrid Hybrid geometries use a combination of the basic geometries. Pastry [Rowstron and Druschel, 2001b] combines the tree and ring geometries in order to achieve more efficiency and flexibility. Peer identifiers are maintained as both the leaves of a binary tree and as points on a one-dimensional circle. In Pastry, the distance between a given pair of nodes is computed in two different ways: the tree

Download sample

Download