It's a network administrator's dream,a system that identifies, isolates and repairs glitches all by itself. As thresholds are reached, servers reallocate resources automatically. When a virus or hacker intrudes, the network responds without involving humans at all, eliminating threats and learning from the incidents so they don't occur again. As a result, the technology reduces overall IT cost of ownership by as much as 50 percent.
Fantasy? Not by a long shot. This is the reality of autonomic computing, a nascent effort designed to make networks configure, optimize, heal and protect themselves. Cutting-edge ISVs have been working on the technology piecemeal for years, writing and selling software programs that automate individual components of overall network management. In October, however, when IBM and Cisco Systems, San Jose, Calif., unveiled a formal partnership to develop an end-to-end solution, the playing field changed completely.
"We stepped back and realized that technology like this could transform network management forever," said Dave Bartlett, director of autonomic computing at IBM, Somers, N.Y. "If you think about the impact that completely self-healing systems could have on computing today, it's pretty mind-boggling."
|NASA plans to enable computers aboard space shuttles to fix themselves.|
The duo's first plan was called the Adaptive Service Framework, and it relied upon a Common Base Event (CBE) format designed to standardize the way the new systems would work. Last month, the twosome was at it again, unveiling their second step toward the ultimate goal. Phase two of the plan revolves around a "generic log adapter" to convert existing log files to the CBE and a "log and trace analyzer" to view and correlate files in the same format. So far, the OASIS standards body has issued no complaints.
While many ISVs appear to be waiting to see how the technology progresses, a handful of developers have grabbed the automation idea by the horns. Solution providers such as Addamark Technologies, San Francisco; Opalis Software, Toronto; and Vision Solutions, Irvine, Calif., all have unveiled plans to adopt IBM's latest problem determination technology. Peregrine Systems, an asset and service management firm in San Diego, already has gotten into the game, using the generic log adapter to convert logs into the CBE, then using the log and trace analyzer to correlate IBM's WebSphere with its ServiceCenter application.
In Seattle, Singlestep Technologies is working with autonomic solutions as well. The company sells an event management product named Unity, which incorporates the CBE to automate certain other aspects of network administration. While this application epitomizes the current state of autonomic computing, Chris Noble, Singlestep's CEO, said that few of the company's customers know it as such, and just think of Unity as a product that helps them free up human resources for other tasks.
"We've never had a customer come to us and say, 'I need an autonomic computing solution,' " Noble said. "That doesn't mean we're not automating things anyway."
Some ISVs are even taking automation to the micro level. Tampa, Fla.-based startup Persystent Technologies, for instance, recently unveiled a product called iNode that detects and self-repairs problem files and registry settings on individual servers and PCs.
Jeff Kruger, Persystent's director of marketing and business development, notes that while this fledgling product might not fit IBM's formal definition of autonomic computing, in trials it has helped network administrators reassign almost all of their staff resources away from PC support.
Then, of course, there are the noncommercial endeavors. At the National Aeronautic Space Administration's Goddard Space Flight Center in Greenbelt, Md., technologists are developing autonomic computing technology that would make IBM's plan for business enterprises seem like the minor leagues. According to Peter Hughes, assistant chief of technology in the center's Information Systems Division, the NASA system is designed to achieve end-to-end system autonomy for future space shuttle flights, enabling computers aboard future missions to fix themselves.
"If a failure occurs in space, we want to be able to rest assured that the system can correct itself without intervention of any kind," Hughes said. "On the business level, you can always throw humans at a problem in a pinch; in space, it's a different situation entirely."
At NASA, Hughes guessed that these kinds of self-healing technologies are years away. At IBM, Bartlett warned it could be even longer before industry-approved autonomic computing becomes reality. Both declined to predict when the technology would arrive formally in the enterprise world, but Bartlett said that IBM and Cisco would most likely release the next round of self-healing standards before April 1.
Until then, as ISVs such as Peregrine and Singlestep continue to develop stand-alone automation technologies, it appears all one can do is dream.
"Autonomic computing isn't going to happen overnight," said Laurie McCabe, vice president at IT research firm Summit Strategies. "When it does happen, though, it's going to be huge."