Policy-Based Autonomic Computing

by Richard Anthony (Editor), Omer F. Rana (Editor), Duncan Johnston-Watt (Editor)

Policy-based computing is one of many techniques used to implement autonomic computer systems. The business function is expressed in terms of workflow (logic + actions) or behavioural rules to be performed in response to certain situations, which is decoupled from the implementation mechanism. The mechanisms are fixed at design time, but the actual run-time behaviour is decided by the policy. This facilitates contextual awareness as the policy may take account of various environmental inputs from sensors and external software systems. Policy-based mechanisms are also very useful in specifying system behaviour -- such as when specifying Service Level Agreements, and subsequently monitoring whether these agreements have been adhered to. Policy-based configuration is highly versatile and generally applicable across a very wide application space. When compared with other autonomics techniques, policies represent one of the lowest risk and lowest cost solutions; because there is relatively low complexity. This book brings together various aspects of using policies in the context of autonomic computing. It contains papers that report novel contributions to policy-based computing within an autonomics context, along with survey articles that highlight research directions in policy languages, tools and frameworks to support the construction and management of autonomic systems.

250 pages, Paperback

First published March 1, 2008