The progress of modern scientific research in high energy physics, astrophysics, biology, earth science, chemistry, medicine, nanotechnology, industry, business and other areas of human activity requires the joint efforts of many organizations processing large volumes of data in a relatively short period of time. This calls for a geographically distributed computing system that can transfer, process and store huge amounts of data.
Grid, a new type of computing infrastructure, (the name is derived by analogy with electrical networks – electric power grid) is intensively developed in today’s world of information technologies. Grid provides global integration of information and computing resources. The essence of Grid initiative is to create a set of standardized services to provide a reliable, compatible, inexpensive and secure access to geographically distributed high-tech information and computing resources – a separate computer, cluster and supercomputing centers, information storage, networks, scientific tools, etc.
The mission of the Internet was to globalize the information exchange, and the World Wide Web standardized the search and the delivery of documents. Grid became the next stage in this chain of revolutionary transformations – standardization and globalization of the use of all types of computer resources.
“A Grid is a system that coordinates resources that are not subject to centralized control using standard, open, general-purpose protocols and interfaces to deliver nontrivial qualities of service.” (Ian Foster)
Grid combines technology, infrastructure and standards. Special software, which enables to provide the resources (computers, data storage, network, and others) in common use, refers to the technology. The infrastructure consists of hardware and services (based on the human and software resources) that must be organized and constantly supported for the sharing of these resources. Finally, the standards define the format and protocols for messaging between services and between services and users, and also the work rules of the grid.
Grid applications include:
- Complex modeling on remote supercomputers;
- Collaborative visualization of very large sets of scientific data;
- Distributed processing to analyze the data;
- Combination of scientific tools with remote computers and data storage systems.
At the moment the main areas of using the grid can be identified as:
- organization of the effective usage of resources for small tasks with the utilization of temporary idle computer recourses;
- distributed supercomputing in solving very large problems, requiring enormous processing resources, memory, etc.;
- calculations involving large amounts of geographically distributed data, for example, in meteorology, astronomy, high-energy physics, medicine, earth sciences;
- collective computing in which users from the various organizations are involved.