Internet System Administration

A. MOTIVATING FACTORS OF THE P2P NETWORK DEVELOPMENT

Peer-to-peer (P2P) applications account for 50% to 90% of overall Internet traffic (Svensson, 2007). Traditionally, the exchange of resources and services between computer systems is done using client-server techniques. However, a system was developed where prohibitively expensive dedicated servers are not required to centrally manage the data that flows between nodes. (Becker and Clement, 2006).

Engle and Khan (2006) also surmised that the development of P2P systems was inspired by the need to “…realise a computing architecture which cannot be taken down by attacking any single point.” This factor motivated the decentralised design of P2P systems, where multiple millions of users are distributed over the Internet and a single point of attack is unable to critically injure the entire system.

In the early 1990s, Napstar popularised P2Ps through the online distribution of media content. Motivated by the widespread popularity of unrestricted (albeit contentious) online distribution of unlicensed, copyrighted music, the architecture was further developed to include support for other distribution applications (Becker and Clement, 2006). Even after the demise of Napster in 2003, the demand for P2P file sharing seemed to increase. Indeed, Napster’s very existence seemed to give birth to a number of other P2Ps, including iMesh, OpenNap, KaZaA, Gnutella and most recently, BitTorrent (Krishnan et al., 2007).

The development of the P2P network architecture, also sought to exploit unused processor space, and to create a vast resource into which multiple clients could tap. Today’s modern personal computers comprise exceptionally fast processors, huge memories and large hard disks, none of which were being fully utilise in the client/server environment (Davies, 2004).

Using current technology, the cost of servicing millions of users on a traditional client/server network would be exorbitant....