Peer to Peer Networks

Recently there has been a substantial amount of research in P2P networks. For
example, P2P network topology has been an area of much interest. Basic peer
networks include random coupling of peers over a transport network such as
Gnutella (http://www.gnutella.com)  and centralized server networks such as
that of Napster (http://www.napster.com) architecture. These networks suffer
from drawbacks such as scalability, lack of search guarantees, and
bottlenecks. Yang and Garcia-Molina discussed super-peer
networks that introduce hierarchy into the network in which super-peers have
additional capabilities and duties in the network that may include indexing the
content of other peers. Queries are broadcasted among super-peers, and these
queries are then forwarded to leaf peers. A network in which a deterministic
topology is maintained and known of by all nodes in the network. Therefore,
nodes at least have an idea of what the network beyond their scope looks like.
They can use this globally available information to reach locally optimal decisions
while routing and broadcasting search messages. Content addressable networks
(CAN) have provided significant improvements for keyword search. If
meta-information on a peer’s content is available, this information can be used
to organize the network in order to route queries more accurately and for more
efficient searching. Similarly, ontologies can be used to bootstrap the P2P
network organization: peers and the content that they provide can be classified
by relating their content to concepts in an ontology or concept hierarchy. The
classification determines, to a certain extent, a peer’s location in the network.
Peers routing queries can use their knowledge of this scheme to route and
broadcast queries efficiently.


Peer network layouts have also combined multiple ideas briefly mentioned
here. In addition, Nejdl proposed a super-peer based layout for
RDF based P2P networks. Similar to content addressable networks,
super-peers index the metadata context that the leaf peers have. Efficient
searching in P2P networks is very important as well. Typically, a P2P
node broadcasts a search request to its neighboring peers who propagate the
request to their peers and so on. However, this can be dramatically improved.
For example, Yang and Garcia-Molina have described techniques to increase
search effectiveness. These include iterative deepening, directed Breadth First
Search, and local indices over the data contained within r-hops from itself.
Ramanathan, Kalogeraki, and Pruyne proposed a mechanism in which peers
monitor which other peers frequently respond successfully to their requests for
information. When a peer is known to frequently provide good results, other
peers attempt to move closer to it in the network by creating a new connection
with that peer. This leads to clusters of peers with similar interests that allow to
limit the depth of searches required to find good results. Nejdl proposed using the
semantic indices contained in super-peers to forward queries more efficiently. Yu
and Singh proposed a vector-reputation scheme for query forwarding and
reorganization of the network. Tang, Xu and Dwarkadas made use of data
semantics in the pSearch project. In order to achieve efficient search, they rely on a
distributed hash table to extend LSI and VSM algorithms for their use in P2P
networks.