Welcome to Thaddeus Eze Friday, October 20 2017 @ 11:17 PM CEST


Research Focus -Autonomic Computing

IBM in 2001 observed in a manifesto [1] that the main obstacle to the future growth of information technology (IT) is a looming software complexity crisis. The innovations in software development have increased exponentially, the sophistication and complexity of system design thereby stretching human capabilities to the limits to install, configure, optimise, maintain and manage these systems. The software is so complex that almost no single person knows everything about the software anymore. Developers cannot say for certain that there’s no bug in their software as the testing on time frame does not allow them to get every bug out. Now people release products with known bugs in them hoping to fix the bugs in subsequent version(s). Another issue is the question about upgrade. When one component of a system is upgraded, what happens to the other components that integrate with it and do they still work or does it result to loss of functionalities? Alongside that is the drive for greater functionality in our software –software that’s smarter and does greater things –that is also driving the software complexity. Autonomic Computing has been chosen as a way forward [2] –the idea of designing computing systems to manage themselves in the same fashion as the biological autonomic nervous system using high-level policy objectives set by human administrators. AC consists of four self-* properties –self-configuring, self-optimising, self-healing and self-protecting.

Autonomic technology is advancing at a high rate, yet there are no universal standards for the technology itself and the design methods used. There are also significant limitations to the way in which these systems are validated, with heavy reliance on traditional design-time techniques, despite the highly dynamic behaviour of these systems in dealing with run-time configuration changes and environmental and context changes. These limitations ultimately undermine the trustability of these systems and are barriers to eventual certification. We strongly believe that certification is critical to achieving the full goal of AC. We have a longer term vision to develop trustworthy and certifiable autonomic systems and hope to progress towards this through defining validation techniques. We propose that one vital step in this chain is to introduce robust techniques by which the systems can be described in universal language, starting with a description of, and means to measure the type and extent of autonomicity (autonomic functionalities) they provide.

[1] Horn Paul, Autonomic computing: IBM perspective on the state of information technology, IBM T.J. Watson Labs, NY, 15th October 2001. Presented at AGENDA 2001, Scottsdale. Available via http://www.research.ibm.com/autonomic/manifesto/autonomic_computing.pdf

[2] J. O. Kephart and D. M. Chess, The vision of autonomic computing, In IEEE Computer, volume 36, pages 41–50, January 2003

Research Interests

• Autonomic and self-managing systems
o Special interest in trustworthiness
• Mobile ad hoc networks
• Wireless sensor technology

Last Updated: Tuesday, December 27 2011 @ 03:02 PM CET| Hits: 33,655 View Printable Version