Peer-to-Peer Background
Peer-to-peer gained popularity with Napster (1999–2001), an online music file sharing service created by Shawn Fanning. Although free copying and distributing of music files by the users led to a copyright violation lawsuit against Napster, and eventually closing of the service, it paved the way for peer-to-peer file-distribution models that came later. Gnutella had its first release in March 2000. It was followed by Fast-Track (used by Kazaa), BitTorrent, WinMX, and GNUnet in March, April, May, and
November of 2001 respectively.
P2P Networks
Internet users that are ready to share their resources become peers and form a network. When a peer in the network has a file (for example, an audio or video file) to share, it makes it available to the rest of the peers. An interested peer can connect itself to the computer where the file is stored and download it. After a peer downloads a file, it can make it available for other peers to download. As more peers join and download that file, more copies of the file become available to the group. Since lists of peers may grow and shrink, the question is how the paradigm keeps track of loyal peers and the location of the files. To answer this question, we first need to divide the P2P networks into two categories: centralized and decentralized.
In a centralized P2P network, the directory system?listing of the peers and what they offer uses the client-server paradigm, but the storing and downloading of the files are done using the peer-to-peer paradigm. For this reason, a centralized P2P network is sometimes referred to as a hybrid P2P network. Napster was an example of a centralized P2P. In this type of network, a peer first registers itself with a central server. The peer then provides its IP address and a list of files it has to share. To avoid system collapse, Napster used several servers for this purpose, but we show only one in Figure below A peer, looking for a particular file, sends a query to a central server. The server searches its directory and responds with the IP addresses of nodes that have a copy of the file. The peer contacts one of the nodes and downloads the file. The directory is constantly updated as nodes join or leave the peer. Centralized networks make the maintenance of the directory simple but have several drawbacks. Accessing the directory can generate huge traffic and slow down the system. The central servers are vulnerable to attack, and if all of them fail, the whole system goes down.
Decentralized Network
A decentralized P2P network does not depend on a centralized directory system. In this model, peers arrange themselves into an overlay network, which is a logical network made on top of the physical network. Depending on how the nodes in the overlay network are linked, a decentralized P2P network is classified as either unstructured or structured.
Unstructured Networks
In an unstructured P2P network, the nodes are linked randomly. A search in an unstructured P2P is not very efficient because a query to find a file must be flooded through the network, which produces significant traffic and still the query may not be resolved. Two examples of this type of network are Gnutella and Freenet.
Gnutella The Gnutella network is an example of a peer-to-peer network that is decentralized but unstructured. It is unstructured in the sense that the directory is randomly distributed between nodes. When node A wants to access an object (such as a file), it contacts one of its neighbors. A neighbor, in this case, is any node whose address is known to node A. Node A sends a query message to the neighbor, node W. The query includes the identity of the object (for example, file name). If node W knows the address of node X, which has the object, it sends a response message that includes the address of node X. Node A now can use the commands defined in a transfer protocol such as HTTP to get a copy of the object from node X. If node W does not know the address of node X, it floods the request from A to all its neighbors. Eventually one of the nodes in the network responds to the query message, and node A can get access to node X. one of the reasons that Gnutella cannot be scaled well is the flooding.
node A needs to know the address of at least one neighbor. This is done at the bootstrap time. The software includes a list of nodes (peers) that node A can record as neighbors. Node A can later use the two messages, called ping and pong, to investigate whether or not a neighbor is still alive.
المادة المعروضة اعلاه هي مدخل الى المحاضرة المرفوعة بواسطة استاذ(ة) المادة . وقد تبدو لك غير متكاملة . حيث يضع استاذ المادة في بعض الاحيان فقط الجزء الاول من المحاضرة من اجل الاطلاع على ما ستقوم بتحميله لاحقا . في نظام التعليم الالكتروني نوفر هذه الخدمة لكي نبقيك على اطلاع حول محتوى الملف الذي ستقوم بتحميله .
|