Three companies—AMD, Oracle and Sun Microsystems—have used three separate approaches to distinguish their products from the rest of the technology pack. While it is tough to guess whether those approaches will pay off in the marketplace, it is good for the technology industry and good for end users to have new choices. The respective approaches are evolution, revolution and innovation.
AMD picked evolution as a way to avoid spending its life in Intels slipstream. Bits still matter in the computing industry, and computing platforms will inevitably move to 64-bit architectures just as they moved from 8-bit to 16-bit to 32-bit. The 64-bit systems provide more than enough headroom for even the most demanding tasks.
But asking customers to leave behind their application legacy is a major request in this period of restricted budgets. Intel is contending the transition on the desktop from 32 bits to 64 bits is further away and will be a leap from one platform to the other. AMD is betting it will be closer in and marked by customers wanting to run both 32- and 64-bit platforms on the same system, preserving their present applications while building new applications that take advantage of the wider word capabilities.
In an interview with eWEEK Executive Editor/News Michael R. Zimmerman, AMD CEO Hector Ruiz contended that the desktop migration to 64-bit on the client side will happen more rapidly than many predict. Of course that is what you would expect the boss of a company introducing a 64-bit client chip to say, but I think there is some evidence to support Ruizs contention. On the enterprise side, the rise of high-bandwidth networks and dispersed development teams really starts to strain the engineering and design applications that traditionally use the computing-intensive tasks. On the consumer side, the increased use of the home network to shuttle video and audio about the household will strain the current client systems.
Oracle is promoting the idea of a revolution in computing based on grid technology. The idea is that a managed network of low-priced systems can outperform, scale faster and be more reliable than a single large system. This is quite a change from the company that built its business on one big database running on big boxes. The one big database is still there, but the database now runs over an ever-changing grid array. Promoting a revolution is a lot harder than advising evolution. You need to be able to prove that products such as grids have the reliability, ease of management and security that are claimed. You are asking customers to change the way they have been operating their IT infrastructures. That is no easy task.
Sun has gone from being a somewhat- arrogant leader in high-end computing to searching for footing in a Linux world running on commodity hardware. To its credit, the company is working at innovating its way out of the current environment. Technology aside, I think the best example so far has been its pricing structure. Pricing and licensing have consistently been a thorny problem for both vendors and customers. Pricing too often becomes so complex that no one can really understand what he or she is paying for. Pricing per server does not make much sense in a world of grids and numerous blade servers being brought into play depending on demand. Pricing by user access becomes meaningless in a world of Internet computing and Web-based applications.
In the end, what someone needed to do was just simplify the pricing issue. Although, as we reported in the Sept. 22 issue of eWEEK, $100 per user per year is not quite as simple as Sun portrayed, the simple pricing plan has a lot of merit. Rather than use pricing as a method to confuse the end user, pricing may actually become a way to compare products. I dont think other vendors can ignore the need to develop pricing programs that corporate users can understand.
Those three factors—evolution, revolution and innovation—have stood the test of time in helping companies succeed in the face of tough economic and market-segment competition. It is good to see those factors are still alive and well.
Discuss this in the eWEEK forum.
Editor in Chief Eric Lundquists e-mail address is firstname.lastname@example.org.