By submitting your email address, you agree to receive emails regarding relevant topic offers from TechTarget and its partners. You can withdraw your consent at any time. Contact TechTarget at 275 Grove Street, Newton, MA.
First of all, we need to understand that blades and bricks aren't mutually exclusive and that a blade is only one component of a larger, high-density server unit, rather like the half-human occupants of a Borg cube in a Star Trek movie. With IBM's latest offering, based on Intel's Xeon processor, we're now up to a possible 84 blades crammed on to a single rack.
But rather than clearing out banks of older servers at a single sweep, the arrival of this tightly stacked blade technology in the server room represents an evolution rather than a replacement for more traditional Big Iron computing, and I would expect to see the two coexist quite happily.
Blades represent a great clustering solution through offering many servers in a single unit but bricks, by adding more components to each blade - a chunkier blade if you like - represents a more interchangeable, scalable solution, which is more appropriate for resource sharing.
It's easy to understand why blades are going to be a hot topic for some time to come. This is, after all, a much cheaper way of adding processing capacity without the extra data centre costs of space and power. Sitting in the middle of an IT recession, businesses are, increasingly, cost and productivity conscious, and Gartner predicts that the market will be worth $3.7bn by 2006.
But any new technology solution carries its own health warning and blades and bricks aren't immune. Standards and compatibility issues still need to be resolved and it's still early days for management software. It's argued that blades aren't really well suited to enterprise applications, although with IBM and, indeed, Sun Microsystems now playing in this space, I would expect many of today's problems and objections to have been resolved by the middle of next year.
So the bricks over blades argument is a little specious as both terms really appear to represent one and the same thing - high-end modular computing. They aren't really very different, in principle, to the multi-function cards in the back of a PC, some being more functional and thicker than others.
Move forward 18 months and large companies will start to consider the server solution they require. Big Iron, the kind of expression one associates with Unisys and its very successful ES7000 or server blades - or bricks - from IBM or Hewlett-Packard or Sun perhaps. Of course, the final decision will be determined by the environment or the application, a solution best suited to one or the other, clustered or enterprise perhaps?
By 2006 I suspect we'll take the choice for granted, choosing the right server for the task in much the same way as we might choose a family car.
What is your view?
Do you think blades will revolutionise the way we use servers? Tell us in an e-mail >> CW360.com reserves the right to edit and publish answers on the Web site. Please state if your answer is not for publication.
Zentelligence Setting the world to rights with the collected thoughts and opinions of the futurist writer, broadcaster and Computer Weekly columnist Simon Moores.