Welcome to Orcmid's Lair, the playground for family connections, pastimes, and scholarly vocation -- the collected professional and recreational work of Dennis E. Hamilton

 

Autonomy, Where Art Thou?

ACM News Service: Self-Managing Systems Not So Self-Evident.  I am somewhat baffled by this lead: "Autonomic computing experts agree their field is still relatively immature, and that vendors still need to build out their product portfolios and educate people about self-managing technology."

IBM architect Jeffrey Frey "said the reality was that enterprise systems and data are constantly in flux, and that management needs to be able to quickly address that fluctuation with rapid and autonomic decision-making."  I think this refers to the question about being able to interoperate systems management across autonomic systems in an enterprise, but I am wondering who trains this, or is trained for it.

I am puzzled about the prospect of autonomic computing in general. I am baffled how we are to manage emergent behavior by monkeying in the plumbing.  Prozac for computers?  At what level of abstraction is the notion of autonomic computation manifest and how does it figure in the way we engage systems in accomplishment of useful work on behalf of enterprise?

Self-Managing Systems Not So Self-Evident.  Clint Boulton's 2004-05-19 InternetNews.com article describes the panel discussion at the first International Conference on Autonomic Computing.

As introduced by IBM in 2001, autonomic computing "is a term used for a network's ability to self-manage, self-heal and self-configure systems -- be they Web services or data center infrastructure -- on the fly to make sure networks suffer as little downtime and require as little maintenance as possible. The idea is to free up software engineers to conduct other critical tasks."  One can see how this emphasis (from IBM's Jeffrey Frey) suggests a reduction in labor-intensive trouble-shooting and repair of mission-critical systems.

One interesting difference among the panelists is the level of granularity and monitoring at which antonomic operations function.  This accounts for differences between IBM's on-demand use of autonomic computing, HP's Adaptive Enterprise strategy, and Microsoft's Dynamic Systems Initiative.

There is lack of any agreement on approach and standardized information that might be used in coordinating and management across the enterprise. The account by Amazon's Jacob Gabrielson starts with reality -- the challenge of maintaining operation of amazon.com and its back-end and plant operations. "He lamented that single points of failure were problems, and said the company uses a combination of 'plug-an-play' infrastructure and proprietary software to help its computers communicate."

It seems that it comes down to basic questions: How does one apply hot fixes and/or roll-backs, how are they set up, and how are they established? How does one confirm the validity of an adaptation/re-configuration maneuver?
posted by Dennis at 5/23/2004 08:07:03 AM