By SCOTT DEWING
Published: July 2003

I’VE BECOME INCREASINGLY SKEPTICAL of anything that’s been dubbed “the next big thing”—especially when it comes to technology. It seems that every technology company promises that their product or solution is “the next big thing” as if the very survival of the human race was somehow dependent upon the next version of Microsoft Windows or the release of Intel’s latest processor. Although these may be the next big thing for these companies (and for their shareholders), they may not mean squat to you and me.

So it was that my built-in, shock-proof skepticism alarm went on high alert recently while reading an article about grid computing that contained the phrase “the next big thing.” Really? What is this grid thing and why is somebody telling me it’s “the next big thing”?

Smelled like marketing to me.

Put simply, grid computing is a distributed computing model in which a large collection of interconnected and heterogeneous computers work together as one big, “virtual” supercomputer. If the term “interconnected” made you think of the Internet, then you’re onto why some folks believe that grid computing really is the next big thing. Some have gone as far as to hail it as the next evolution of the Internet.

According to an IBM white paper on grid computing, “The standardization of communications between heterogeneous systems created the Internet explosion. The emerging standardization for sharing resources, along with the availability of higher bandwidth, are driving a possibly equally large evolutionary step in grid computing.”

There are millions of computers connected to the Internet. Most of the time these computers are sitting idle. Some studies have concluded that desktop computers are busy less than 5% of the time, which in and of itself is an interesting commentary on what the users of those computers are doing the other 95% of the time.

According to IBM, “Grid computing provides a framework for exploiting these underutilized resources and thus has the possibility of substantially increasing efficiency of resource usage.” For example, scientific research projects and financial analysis that require massive number crunching could utilize a “grid” of interconnected computers for processing data rather than having to invest in building individual (and very expensive) supercomputer infrastructures.

According to a recent article in The Wall Street Journal, financial firms are, “wrestling with bloated transaction-processing infrastructures and distributed environments that are costly to maintain. Estimates are that only 15 to 20 percent of their compute cycles are being utilized. These facts are causing industry insiders to ask: How about migrating off heavy iron and running the compute-intensive risk calculations on commodity hardware, whose price is one-tenth the cost? And why not exploit those unused processing cycles and optimize the performance of applications by allocating jobs to idle machines?”

Some financial institutions have indeed begun to turn to grid computing to increase their computing power while decreasing their hardware costs. Charles Schwab, for example, recently worked with IBM to speed up its portfolio-rebalancing software by distributing the processing needs amongst its many servers. Like all online stock trading firms coping with the Internet trading boom, Schwab had to invest heavily in computing infrastructure to keep up with demand during peak trading hours. During off-peak hours, however, all of that hardware just hums along doing little to no processing. With grid computing, Schwab was able to utilize these servers to speed up the portfolio-rebalancing software processing. The result: what used to take 4 minutes, now takes 15 seconds.

The dream of grid computing is to use those millions of computers across the Internet. However, don’t expect Schwab or any other company to be using your computer anytime soon for processing sensitive data. Perhaps the biggest hurdle to grid computing becoming a standard is, of course, security. Data security is not “the next big thing”—it is the big thing when it comes to the Internet and will remain so, perhaps indefinitely.

“If you were the CEO of a big company, would you be comfortable sending your intellectual property across the Internet to a third party who might be working with your competition?” asked Peter Jeffcock, a product manager for Sun Microsystems, in a recent interview.

Good point. According to Sun, adoption of global grids has been slowed because of security concerns and a lack of standards. Sun is attempting to address the security issue of grid computing with a product called the Sun Grid Engine, which builds security layers into the grid framework. Currently, however, Grid Engine is only being used for building intercompany, and internal, private grids.

Sun and IBM are the big companies at the forefront of the grid computing initiative. I suspect it will only be a short time before Microsoft percolates to the top of the marketing cauldron with something like the Microsoft Active Grid Internet Client (or MAGIC) and declares it, for better or for worse, “the next big thing.”