Grid Computing
| The same trends that brought about P2P computing cheap computers, bandwidth, and idle processors also contributed to the emergence of a related concept in distributed computing called grid computing. One of the main goals of peer-to-peer computing is to take advantage of computing power at the edge of the network by direct interaction between these devices (without the intermediary of a central server), as exemplified by the typical P2P usage for file searches. This goal emphasizes the concept of individual peers as collaborating computing entities, and the concept of peer independence. Grid computing has some goals similar to P2P, but with an emphasis on large-scale resource sharing for the purpose of high-performance computing. Whereas the problems tackled by current P2P applications do not require large resource commitments from the peer nodes, grid computing technologies have emerged in response to the so-called "grid problem," in which large-scale computing resources are needed, and those resources cross organizational boundaries. The grid thus refers to an infrastructure that enables the integrated, collaborative use of high-end computers, networks, databases, and scientific instruments owned and managed by multiple organizations. Grid applications often involve large amounts of data and/or computing, and often require secure resource sharing across organizational boundaries that are not easily handled by today's Internet and Web infrastructures. This concept of sharing computing resources across virtual organizations raised a large number of issues that traditional distributed computing did not address, but that were very close to some of the issues that the P2P model faces: how to structure fine-grained access control over resources; taking care of local and global policies; and how to agree on quality of service, scheduling, and co-allocation. In response, research groups have been developing open architectures and standards to solve some of these problems. One of the most advanced from the viewpoints of acceptance and practicality is the Globus Project (http://www.globus.org). The Globus Project is a research and development project focused on enabling the application of grid concepts to scientific and engineering computing. One of the main deliverables of the project has been the Globus Toolkit, a set of services and software libraries that supports the development of computing grids and grid applications. The toolkit has four major components: The Globus Resource Allocation Manager (GRAM) provides resource allocation and process creation, monitoring, and management services; the Grid Security Infrastructure (GSI) provides a single-sign-on, run-anywhere authentication service; the Metacomputing Directory Service (MDS) is an extensible grid information service that combines data discovery mechanisms with the Lightweight Directory Access Protocol (LDAP); and the Global Access to Secondary Storage (GASS) implements a variety of automatic and programmer-managed data movement and data access strategies. Several large software and hardware companies have now joined the grid computing movement and are contributing to the project. As a result, we have seen the rise of the Open Grid Services Architecture (OGSA) and related technologies for locating and managing grid resources. You can read more at http://www.globus.org/ogsa. Grid Computing ExamplesThere are currently several good examples of grid computing applications. A well known, if relatively simple, example of shared computing resources is the SETI@home project, managed by the Space Sciences Laboratory of the University of California, Berkeley. SETI, the Search for Extraterrestrial Intelligence, is a scientific effort seeking to determine if there is intelligent life outside Earth (http://setiathome.ssl.berkeley.edu). The SETI@home project is trying to detect signals of extraterrestrial intelligence by scanning a 2.5MHz band of radio frequencies and processing them to look for narrowband signals. The computations involved would overwhelm the largest supercomputers in existence. In order to solve that problem, the project leverages the idle processing power of millions of computers connected to the Internet, whose owners contribute willingly to the project. You can contribute by downloading a client program that runs as a screensaver and processes data for the project while your computer is idle. On a periodic basis, this client connects to the SETI@home servers to upload results and download new data for the next computation. Although it is a type of grid computing, the SETI@home computations are well structured and involve all the computers asynchronously running the same client code, making it a relatively simple application of grid computing that does not require the use of complex software such as the Globus Toolkit. Instead, the SETI project handles the data distribution and aggregation from central data and task servers. More complex and technically significant examples of grid computing include the following:
|
