Companies today need a competitive edge to stay alive. Grid computing provides a way to harness all available computing resources to meet business objectives and compete to win. It allows companies to do more at a much lower cost by providing resources to meet overall demand.
Grid Computing Features and Benefits |
|
|
Time to market |
|
|
Increase quality/reliability and add features |
|
|
Harness all available resources |
|
|
Elastic, flexible, scalable compute environment |
|
|
Ability to use lower cost commodity hardware |
|
|
Do things you couldn't do before |
|
|
|
Grid computing can be defined as an expandable, scalable set of resources applied to solve a single or a set of problems - usually a scientific or technical problem that requires a large number of computer processing cycles. This can be seen in small and large companies alike as they design integrated circuits, do drug discovery, molecular modeling, financial simulations, or any number of other compute intensive calculations.
Grid is an architecture that enables dynamic allocation of resources to varying workloads in accordance with business needs. Users do not care where the compute cycles or data reside - it is delivered to them quickly, efficiently, and seamlessly.
Grids are built with low cost modular components, so you can start small and preserve your investment as your business grows.
The roots of Grid computing came from the university and scientific communities in the 1980s and early 1990s. It has grown from a viable option to a necessity for companies in High Performance Technical Computing (HPTC). As the business use of Grids has grown, so has the investment in reliable bullet-proof software. Because of the complete set of reliable products and services, Grids are mainstream in HPTC, and are now being adopted in markets using commercial database applications from Oracle, SAP and others.
The basic components of a Grid include: |
|
Compute Grid |
|
|
Data Grid |
|
|
Access Grid |
|
|
Commercial Grid |
|
|
|
|
Compute Grid is a set of hardware and software that forms a compute infrastructure where compute intensive batch applications are run. The batch jobs could be short jobs that take a few seconds or minutes to complete, to very long jobs that can take hours or days. These jobs run on the compute grid that is designed and tuned to provide the fastest possible turnaround given the application set. The compute components could be confined to the network of computers within a corporation or it can be a public collaboration.
Data Grid is the storage component of a grid environment. Scientific and engineering applications require access to large amounts of data, and often this data is widely distributed. A data grid provides seamless access to the local or remote data required to complete compute intensive calculations.
Access Grid is the set of hardware and software used to submit compute jobs. The hardware could be any windows or unix based computer with network access to the compute and data grids. One prominent grid portal software product is EnginFrame from NICE. Grid portal software is not required to submit compute jobs or access the grid, but it does simplify the process.
Commercial Grids are focused on database applications. Oracle, SAP and other database companies have introduced solutions that allow users to take advantage of Grid computing environments for commercial applications. This growing trend is the next evolution in Grid Computing. For more information on Oracle's Grid solutions, go to http://www.oracle.com/technology/products/oracle9i/grid_computing/index.html. For information on the Sun/SAP solution, go to http://www.sun.com/service/n1gridsap/
Grid computing is the information technology infrastructure of the future, and it is here today!
For more information, please email info@gridgateway.com
|