Grid computing gets down to business

The grid virtualises underlying technology so that the user sees what is essentially one large computer. Grid computing is a business thought and the end-game is to enable organisations to get more out of their investments in people and technology, says R Dhamodaran

In the wake of the rise and fall of the dotcom era, new technologies that promise the moon and stars are often regarded with a healthy dose of scepticism.

Recently there’s been a lot of discussion—even some measure of hype—about grid computing. Many business leaders are asking if a grid is really a new approach to computing or just the Next Big Thing.

Origins in academia

Grid computing, like the Internet, has its origins in research and academia, and has moved into the commercial mainstream. However, before it can become an integral part of a company’s business and IT strategy, we need to be clear what grid computing really is, and the benefits it can bring to commercial businesses.

Grid myths

Most people tend to think of a computing grid as a network of geographically-distributed systems hooked together to tackle a single, computing-intensive task, a kind of virtual supercomputer. The example often cited from the realm of science and research is the SETI@home project, which uses the excess processing power on PCs around the globe to examine radio signals for evidence of extraterrestrial life.

The idea of a grid as just a distributed supercomputer focused on a single problem is light-years off course.

It’s true that many organisations are using grid technologies to build virtual supercomputers that help them solve complex computational problems. This is one way—but by no means the only way— businesses can benefit from grid technology.

Enterprises have made significant investments in computing and data assets, much of which sits idle 80 to 90 percent of the time, so they are looking to get more value out of these investments.

Untapped capacity

When computing resources are grid-enabled, untapped capacity becomes available on demand, similar to the way processing power in a mainframe computer can be directed on-the-fly to where it’s needed.

Think of grid computing as the evolution of distributed computing. It’s not just about providing access to massive processing power. The real business value of a grid lies in the ability to pin together disparate and widely-dispersed computing and data resources across an organisation in new ways. Companies can harness these previously-untapped assets to accelerate business, analytical and scientific processes, improve collaboration, and provide employees on-demand access to vast IT resources.

Virtual mainframe

Perhaps a more accurate way to look at grid computing is to visualise all the systems in an enterprise working together to create a powerful virtual mainframe, giving an organisation unprecedented access and control over computing and data resources, not only within the network but also across the enterprise. The grid virtualises the underlying technology, so the user essentially sees one large computer.

The majority of IT systems are distributed midrange and desktop systems based on different architectures from multiple vendors, and all too often, these systems don’t work and play well together.

What if a company could connect these diverse, far-flung computing and data resources, unleash their collective processing power, and manage them like a single large mainframe? This is possible with grid computing.

They typically begin by grid-enabling key workloads—such as business analytics or engineering and design applications—and expand their grid initiatives based on the success of these initial projects.

Real world grids

Charles Schwab, for example, recently grid-enabled an existing wealth-management application, reducing processing time from more than four minutes to just 15 seconds. This improvement will allow a Schwab representative to respond to a customer’s question on the phone in seconds rather than calling back later. Now Schwab is looking to expand grid research into other areas of its business.

Similarly, using grid technologies, RBC Insurance was able to reduce by 75 percent the time spent on job scheduling and by 97 percent the time Similarly, using grid technology, RBC Insurance was able to reduce the time spent on job scheduling by 75 percent, and the time spent on processing an actuarial application by 97 percent, freeing up highly-skilled actuaries to focus on important customer issues and analytics. The company is looking to expand the solution across additional applications and business units to improve efficiencies and raise customer satisfaction.

Royal Dutch Shell created a grid infrastructure for its seismic interpretation applications that cuts the time needed to process seismic data while improving the quality of the data, thus allowing the company’s employees to focus upon key problems.

Grid computing is real, and it’s here today.

The author is vice president & country executive, Software Group & Developer Relations, IBM India. He can be reached at rdhamod@in.ibm.com