Home // International Journal On Advances in Intelligent Systems, volume 5, numbers 3 and 4, 2012 // View article
Authors:
Haffiz Shuaib
Richard Anthony
Keywords: Autonomic computing systems; Certification; Performance; Verification; Measurement;
Abstract:
Autonomic computing systems are a promising technology for bending the cost curve associated with information and communication technology (ICT) service management and for aiding the growth and evolution of complex computing systems. Indeed, this has motivated a significant amount of research. However, a central plank to achieving fully-fledged autonomic computing systems is missing i.e., the ability to certify these systems. The certification process will provide a basis; for assessing the quality of autonomic systems with similar functionalities, for assessing the current capability of the system and its suitability to the problem, to assess the impact of a certified component on a system and to resolve legal liability, if the autonomic computing systems were to fail. In this second part of a two-part paper, several steps to rate or certify autonomic computing systems within the context of the targeted application domain are proposed. In the first instance, the autonomic manager architecture proposed in the first part of this work is associated with indices that indicate how mature an autonomic machine is. The maturity index, the layer configuration of the machine and the implemented autonomic self-management properties are used to derive a mathematical expression that describes the machine in qualitative terms. These qualitative metrics in turn point to what quantitative measures or performance characteristics can be obtained from the machine under an evaluation scenario. The proposed quantitative metrics are based on the International Standard Organization's software quality specification i.e., ISO/IEC 9126. Using the software engineering standard for product evaluation i.e., ISO/IEC 14598-4, the four steps for certifying an autonomic computing system are outlined. Finally, an Ant Colony Optimization (ACO) application called Path Finder (PF) is used to demonstrate the proposals in this work.
Pages: 384 to 399
Copyright: Copyright (c) to authors, 2012. Used with permission.
Publication date: December 31, 2012
Published in: journal
ISSN: 1942-2679