Special report details autonomic computing's journey
In "Autonomic Computing: Vision vs. Reality," Michael Biddick starts by noting
Five years ago, IBM's Paul Horn articulated a new way of thinking about Information Technology ... we examine how far the technology of autonomic computing has come -- and where it's headed.
The author goes on to detail the five critical components the industry needs to accept that define automated computing:
- The system must anticipate the best resources to fill a need without involving the user.
- The system must continuously optimize itself.
- The system must have a level of self-awareness that lets it understand its own components.
- The system must have self-healing capabilities so that rountine or catastrophic performance degradations, outages, and security threats have minimal impact on users.
- The system must be able to interact with its external surroundings and be based on open standards.
He then discusses some practical goals a company should be headed towards -- automating business/operational processes in regards to monitoring, developing a service-focused IT strategy that scales across organizational units, off-loading the error-prone manual repetitive tasks via automated workflows, using predefined resolution steps for common problems, and providing standardized integration points for such critical systems as provisioning.
Biddick goes on to detail available tools and provides a pretty good checklist of questions to ask before committing to a toolset.