Literature Review of Routing in Peer-to-Peer Networks

Literature Review of Routing in Peer-to-Peer Networks
13/04/2011 Comments Off on Literature Review of Routing in Peer-to-Peer Networks Academic Papers on Information Technology,Sample Academic Papers admin

A form of P2P networks is the Overlay network. The Overlay networks work by constructing a virtual network topology on top of the physical links between the nodes in the network.  The problems in Overlay networks is that their network topology changes all the time, since, the node join and leave the network at random and the average uptime for each individual node is quite low. That is, even if a route is established, there is not guarantee on the time limit the route will be valid for (Wang, Harrell, Hu and Xia 2001). Hence the designers of routing algorithms face quite a few challenges, some of which are Scalability, Complexity and Anonymity. Scalability is the measure of performance of system as the number of nodes or messages increase. Complexity is defined as the number of steps that a packet needs to travel from source to destination in the worst case scenario. Anonymity is a problem if a network is designed to provide anonymity and it needs to be solved at routing level. Over the years, a number of routing standards have been released in order to provide better routing capabilities to the routers from these perspectives (Buford, Yu and Lua 2008).

Gnutella Routing

Gnutella offered the simplest way of routing in P2P networks. The routing concepts employed in Gnutella-based networks were simple. In order to join a network, a node had to know the address of at least one node on the network beforehand.  Once the connection between these two nodes was established, the new node then broadcasted a ping to find the addresses of other nodes on the network.  Each node on the network maintained connections to multiple nodes. For a node to search on the network, the node sent out a query message to each node it was connected to. These nodes then forward the message to the node with the resource was found.  The time-to-live counter was used in the network to control the number of nodes that were queried (Park, Yang, Park, Kang and Choi 2008).

The routing style adopted in Gnutella networks worked well for small and medium size networks; however, Portmann, Sookavatana, Ardon, and Seneviratne (2001, pp.263) have shown that the cost of searching in a Gnutella-based network increase super-linearly as the number of nodes increased.  As the number of nodes saturates, the network has a tendency to become fragmented.  The process of searching in Gnutella has an exponential complexity. This problem in Gnutella network, lead to further research and development of a more efficient network which included semantic routing and distributed hash tables. These algorithms not only provide less search complexity but an added advantage is the lack of redundant overhead.

Distributed Hash Tables (DHT)

Distributed Hash Tables (DHT) algorithms have found their usefulness in sharing files and data on P2P Networks. DTH algorithms make use of a hashing function. This hashing function takes a variable string as an input and outputs a number generated from the string, called Hash. All files and identifiers in a network using DHT are hashed and their location is stored in a huge hash table which is distributed across the participating nodes.  Chord, an algorithm developed by MIT is an example of DHT Algorithm (Park, Yang, Park, Kang and Choi 2008).

Please go to the order form to order essays, research papers, term papers, thesis, dissertation, case study, assignments on this essay topic.

Related Essays, Research Papers, Term Papers, Thesis, Dissertation, Case Study, Assignments entries.


About The Academic Paper