Grid computing accomplishes its work using computer resources across several administrative domains to meet common objectives. Some applications of grid computing have their own user interface. In such instances, the connection is controlled by the computer operator, such as peer to peer (P2P) file sharing networks. Others are less noticeable background tasks, such as processing of scientific data during a computer's idle time.
Grid computing is often called distributed computing because the project files are distributed among several different nodes. The node is comprised of a single, Internet-connected machine loosely coupled to a central computer that may be hundreds of miles away. This central machine is connected to hundreds, even thousands, of other nodes, all of which are receiving packets, processing data, and sending requests. Depending upon the grid structure, the individual nodes may or may not be able to communicate among themselves.
One of the primary benefits gained by applications of grid computing is the loose coupling of nodes to the central computer to create a very powerful virtual machine. This machine is, essentially, a super-computer that can process data at a much faster rate than any of the individual computers on the grid. Such super-computer grid technology has been effectively used to study and process data related to earthquakes, weather, and even the possibility of life on other planets.
Another advantage of using this type of computing is the two-way communication between the node and the central computer. The grid computing process was originally designed with the idea that the Internet should be more like a power grid. The power grid system sends and receives information from individual points to trace usage statistics. This same communication channel principle makes it possible for applications of grid computing processes to be designed for economically priced Internet services based on usage amounts.
Among the other applications of grid computing, this programming structure can also be useful in forming P2P file sharing networks. In networks of this kind, people from all corners of the globe can share data, including audio, video, and text files. The grid computing system allows the node user to find the desired files using a search function coordinated by the central system and other processing points anywhere on the network.
The drawback to this type of service lies in the node connections. Since this network involves a diverse group of Internet users, the various nodes may be connected by different means. As a result, some users may have a very slow upload connection or an unreliable connection to the network. If a user disconnects the computer with the hosted files before the download is finished, the file may not be available until the machine is brought back online.