Cognitive Networks: Towards Self-Aware Networks

2.3. Advances in Autonomic Computing – Autonomic Networking

This section will examine autonomic networking, which is a new form of autonomic computing that was developed to specifically address the needs of embedding increased intelligence and decision making in networks, network devices and networked applications.

2.3.1. A New Management Approach

The name 'autonomic' was deliberately chosen by IBM to invite comparisons to biological mechanisms. The vast majority of past and current work on autonomic computing focuses on IT issues, such as host systems, storage and database query performance. However, networks and networked devices and applications have different needs and operate under different assumptions than IT issues. Autonomic networking focuses on the application of autonomic principles to make networks and networked devices and applications more intelligent, primarily by enabling them to make decisions without having to consult with a human administrator or user.

The motivation behind autonomic networking is simple: identify those functions that can be done without human intervention to reduce the dependence on skilled resources. This not only lowers operational costs, but also enables these valuable resources to be used for more complex tasks. It also enables the network to respond faster and more accurately to the changing needs of users and businesses. The left of Figure 2.2 shows a human body, while the picture on the right is a UML model of a simple (honest! ) application. ...