Logo Rob Buckley – Freelance Journalist and Editor

Autonomic Systems

Autonomic Systems

Computers are becoming harder and more expensive to maintain. Is the solution to let them look after themselves?

Page 1 | Page 2 | All 2 Pages

Computers are difficult to use and expensive to maintain, a fact to which any finance director will attest. But Paul Horn, senior vice president of research at IBM, wants to change that. In October 2001, he laid out IBM's plans to develop 'autonomic' computers – systems that can look after themselves and that automatically adapt to the tasks required of them.

The project is modelled on the human autonomic nervous system, which regulates and repairs the body, responding to changing conditions without any conscious effort on our part. Horn believes that we need similarly autonomic computer systems.

“Building an autonomic computing network is not optional. In fact, the future of the Internet and ebusiness depends on it,” he says. Horn also argues that, if the current rate of expansion of digital technology continues, there will soon not be enough people to keep the world's computer systems running.

As well as funding 50 university research programmes into autonomic computing over the next five years, Horn is reorganising IBM's research division to include two new vice presidents to coordinate the company's own hardware and software-related efforts. He is also committing 25% of the server group's $2 billion (€2.3bn) annual R&D budget to autonomic projects. “The autonomic concept will encompass an enormous amount of what we do,” says Horn. “It is a unifying theme across all of our labs. It is an overarching vision.”

One of IBM's early autonomic efforts is eLiza, a self-healing and dynamic workload management system announced in March 2001 and scheduled to first appear in the p690 series of Unix servers that was announced in October 2001.

Through the use of internal sensors, it will monitor component health and automatically reallocate memory on the fly.

IBM says it will add more features to eLiza every six months so that, by 2003, eLiza-enabled servers will offer end-to-end automation, authentication and performance management of distributed applications.

Another project is Oceano, a software-based system for groups of servers that need to re-task themselves as user requirements change. For instance, an Oceano-enabled group of web servers could automatically reallocate disk-drive space and processing power between different web sites, according to site traffic levels.

It all sounds too good to be true and, for the time being at least, it is. Analyst group Gartner says that, right now, autonomic computing is little more than a clever combination of PR and brand differentiation. “eLiza is astute PR as it enables IBM to set itself apart in the 'vision' wars from competitors such as Sun Microsystems, HP and Microsoft,” argue Mike Chuba and Thomas Bittman, two of Gartner's server analysts. They also warn that the grand nature of IBM's autonomic vision could end up crippling its efforts.

“IBM Research needs to set specific goals and avoid the risk of getting stuck in the morass of trying to deliver the ultimate heterogeneous solution while smaller, more nimble competitors beat it to the punch with partial solutions,” they say.

Page 1 | Page 2 | All 2 Pages

Interested in commissioning a similar article? Please contact me to discuss details. Alternatively, return to the main gallery or search for another article: