Decision Making and Autonomic Computing in Cloud Computing and Virtualisation-based Environments

Project description

The principles of 'Autonomic Computing' are used to formulate a high-level strategy that could be used to build systems that can change their behaviour at run-time. Such systems are essential to cope with modern parallel/distributed environments where the status of the resources may change quickly in unpredictable ways. For example, allocation of work onto resources following a certain approach may no longer be a good option if the load of these resources (or the network connecting them) changes. A key problem with autonomic computing is how to implement solutions that propose appropriate actions to optimize the behaviour of a system because of such dynamic changes). To do this, a well-known mechanism is control theory, however, it has limitations and doesn't apply to all situations. Other mechanisms are rule-based approaches, heuristics, and many more. An interesting project could combine these approaches and try to build a system that can choose an appropriate and good decision-making mechanism especially in the context of addressing trade-offs in Cloud Computing and related environments

▲ Up to the top