Grid computing is being used widely in science and academia, but only a limited number of database professionals (programmers, database administrators, system administrators) make a living in these circles. How will the grid affect those of us who are involved in industry? What problems can it solve for us?
While there is no extensive use in industry yet, grid computing is making inroads and gaining popularity. Pioneer projects in industry are solving real-world problems and answering business questions for many companies with in-place databases and custom installed databases for support. Many such organizations use a combination of SQL Server on Windows and Oracle on Linux as their data infrastructure of choice.
For more details on the benefits gleaned by the businesses in the following examplesas well as on the part that IBM played in their implementationsee IBM's grid computing success stories.
Financial trading companies are finding grid usage of data stored in databases to be helpful in providing analysis of complex financial products. They are able to perform both real-time and historic trading analytics, determining trends and stock behavior over time, and doing it faster than ever before. This speed, along with more reliable results, gives companies taking advantage of this new technology not only more robust wealth management for themselves and their clients, but enhanced competitiveness and agility in the trading markets. For example, IBM's grid computing solutions have helped to give Charles Schwab's existing wealth management application higher degrees of speed and far faster response times, allowing them to increase customer satisfaction.
Grid computing is helping insurance companies to reduce processing time for financial risk calculations, allowing them to arrive at answers and find trends more quickly, and with more accurate results. These calculations provide actuaries with more timely access to the valuation of highly complex risk-analysis scenarios. For example, Nippon Life Insurance and RBC Insurance have each been able to reduce processing time considerably by implementing an IBM solution. Nippon has been able to cut financial risk calculations, often from nearly 10 hours to just over 45 minutes, say IBM officials.
Automotive and Aerospace Industries
Grid technology and the databases that it accesses assist the automotive and aerospace industries in engineering, design collaboration, customer relationship management (CRM), and marketing. It allows them to cut their analysis and simulation time while improving the speed and reliability of output. The European Aeronautic Defense & Space Company (Airbus included) is using grid technology for system validation and engineering analysis as well as distributed training, and sees many opportunities as well as challenges for the future of this emerging paradigm.
Medicine, Life Science, and Pharmacology
The grid can help pharmaceutical companies to perform calculations that reduce time to drug development, or provide unified data access to medical researchers involved in the discovery of new treatments and cures. As just one example, Oxford University is currently using grid computing to pool and provide access to data on breast cancer treatment. Information is made available on treatment, diagnosis, and screening to researchers, medical professionals, and patients.
Governments can use the grid to maximize data and computing resources and assets across departments through sets of unified data and file interfaces. This means that any piece of data would only need to be stored once (minimizing error), and can be retrieved by anyone with the authority to see that data, anywhere in the system.
Scientists and engineers throughout NASA are using grid technology via the Information Power Grid (IPG), allowing them to access and manage computing, data, instrumentation, and human resources that would otherwise be inaccessible. (See this page of the NASA web site for case studies, projects, and further information.)
The petroleum industry is processing seismic data in a timelier manner with grid technology, while providing better, more reliable output. This means that more time can be spent making use of those result sets, and less time spent waiting for them. Anadarko Petroleum Corporation (details available at Grid Today) and Royal Dutch Shell (another IBM success story) have both implemented grid solutions that allow them to analyze seismic information in the exploration for new, untapped sources of petroleum.
How can the grid help your industry? It's tremendously expensive for companies to constantly maintain peak processing power, when computers are idle more than 50% of the time. By using a grid structure with disparate and distributed databases, adding and removing CPUs and database power driven solely by peaks and valleys in demand, companies could save a significant amount of money and resources by redistributing processing power from underutilized servers to overtaxed servers during peak times, potentially lowering the need for hardware and software resources over time. Programming and administration in the face of this new paradigm will be an adventure and a challenge.