Computer Science 101 Final Paper: In 400 words or fewer, define the following terms while accurately and concisely describing how they differ from one another: Grid Computing, Utility Computing, On-Demand Computing, Data Centre-Virtualisation, and Clustering. (Points will be deducted for marketing terminology or other flowery language.)
Good news for Grid computing. The researchers who invented the original concept and did much of the groundbreaking work have launched a company to provide Grid support and services. This will speed the transition of these powerful ideas from academia into the commercial and enterprise arena: a welcome sign of progress. Yet the commercialisation of the technology comes at a price.
Grid computing has become the latest term in the IT lexicon to be hijacked by sales and marketing. This is typical of many in the vendor community who often use the momentum around emerging technologies to reinvigorate tired products and services. Academic terms are often dry and difficult but they generally have a very specific definition. Yet when a concept becomes fashionable enough to include in marketing campaigns, precisely originated terms slowly degrade into fluffy jargon.
But while this bastardisation of language may lend some short-term spin to a new product or service it damages the longer term potential of a technology to take hold. Grid projects are expensive to implement and raise new problems in management and security, but have great potential for long-term benefits with improved server utilisation, reliability and flexibility. However, presenting a business case to the board for such a project has become much harder due to the marketing chaff thrown out by the IT vendors.
Oracle adopted the grid label for version 10g of its database software. It's nothing of the sort: it just splits the database across a small but tightly linked handful of servers. Sun's grid vision is a global pool of computing power that people pay to use, in the same manner they pay to pull power off today's electrical grid. And HP sees grid computing as a component of its Adaptive Enterprise effort to make corporate computing systems more flexible.
Terminology is just as important as the technology it describes. The history of our industry is littered with examples of fundamentally sound software and hardware concepts that have failed to gain acceptance because of poor understanding or marketing misdirection. Standards bodies and industry groups such as the Organisation for the Advancement of Structured Information Standards, and even the UN, need to tackle the issue. Emerging technologies must be nurtured and protected from the overzealous vagaries of marketers if they are to deliver on their academic potential.