| Grid computing as we understand it today did not exist is any formal sense five years ago. Since then, it has evolved rapidly and is now an important enabling technology for autonomic computing, among other things. Grid has evolved from being considered a small niche-type academic application technology to a centerpiece of IBM's on demand strategy. All the components of grid computing are in place; processors, storage, networks, and other resources that exist today can be configured to make a grid. Grid computing has been defined as just an extension of distributed computing; however this is not the case. Rather it can be said that it is an evolution of distributed computing. It has more potential and takes us beyond distributed computing to the next level of sophistication. Simply put, a grid is a powerful, virtual computer system. It consists of multiple processors in a network sharing resources such as software, storage, data, files, printing, and many other peripherals or resources. The term grid was chosen for its similarity to the pervasive electricity power supply industry. The analogy is correct, as both provide services where required. The basic element of a grid is its computing power, connected by a fast network. Multiple machines from different vendors can be connected together in the grid. In the grid, they do not have to be from the same manufacturer or vendor. They can be geographically dispersed at multiple locations. They can vary in processor speed, memory size, operating system, architecture, types of software and many other factors. Grid computing enables the virtualization of distributed computing and data resources such as processing, network bandwidth, and storage capacity to create a single system, granting users and applications seamless access to vast IT capabilities. Scalability and speed are important factors in grid computing and a measure of how effective the grid can be. Applications normally run and execute on a single computer. If an application can be spilt and designated to other computers in the grid for execution, it should be faster. For example, if an application or job can be run on two machines with equal-sized processors, the application should finish in half the time. If 10 processors are available, then execution should be completed in one tenth of the time, and so on. Scalability will be an important function for autonomic computing management in a grid network. A company with 600 grid-enabled desktop PCs can utilize all of them together as one computing platform—suddenly providing it with enough computing capacity to go head to head with many of the world's largest supercomputers. Additional resources, such as applications, storage, bandwidth and data, are also made available. At its core, grid computing is based on an open set of standards and protocols—Open Grid Services Architecture (OGSA)—that enable communication across heterogeneous, geographically dispersed environments. With grid computing, organizations can optimize computing and data resources, pool them for large capacity workloads, share them across networks, and enable collaboration. |
