Alberto E. Schaeffer-Filho
|Student:||Alberto E. Schaeffer-Filho|
|Title:||A Self-Management Framework for Pervasive Systems|
|Affiliation:||Imperial College London|
PhD project description
Management in pervasive systems cannot rely on human intervention or centralised decision-making functions. The former because pervasive devices must be usable by non technically savvy users. The latter because pervasive devices are mobile and cannot refer to centralised management applications for re-configuration and adaptation. Pervasive systems must be autonomous and continuously adapt to changes in their environment or in their usage requirements. They must therefore be self-managing with local decision making and feedback control to enable seamless adaptation. In essence, this is the proposition of autonomic computing. Previous work has introduced the concept of a Self-Managed Cell (SMC) as a building block for realising pervasive applications. An SMC consists of a set of hardware and software components which form an autonomous administrative domain. SMCs implement a policy-driven feedback control-loop that determines which management actions should be performed in response to events of interest (e.g. device failures or context changes). However, to manage the complexity and scale of pervasive environments it is necessary to define how such environments are structured into autonomous units which interact and collaborate with each other. Although Self-Managed Cells are autonomous elements, in order to accomplish their goals they are required to interact with each other in complex ways, federate or compose into larger structures. For example, a body-area network monitoring a patient's health may comprise “smart” sensors and complex diagnosis devices that are SMCs in their own right. In the same way, an SMC controlling a room will be aggregated under the control of a house SMC and autonomous unmanned vehicles may be aggregated into fleets with a common mission. This thesis proposes a framework for supporting “cross-SMC” interactions that enable collaborations between SMCs in either peer-to-peer or compositional settings, or more complex communities of SMCs. Such interactions permit realising scalable pervasive environments in which SMCs can aggregate into larger structures and engage in ad-hoc peer-to-peer collaborations. This thesis investigates the abstractions required for supporting interactions and federations between SMCs and the main design decisions and architectural choices.
- A. E. Schaeffer-Filho, E. Lupu, N. Dulay, S. L. Keoh, K. Twidle, M. Sloman, S. Heeps, S. Strowes, J. Sventek. Towards Supporting Interactions between Self-Managed Cells. In: 1st International Conference on Self-Adaptive and Self-Organizing Systems (SASO 2007), Boston, USA. IEEE Computer Society, July, 2007. p. 224-233.
- A. E. Schaeffer-Filho, E. Lupu. Abstractions to Support Interactions between Self-Managed Cells. In: Autonomous Infrastructure, Management and Security (AIMS 2007) doctoral student workshop, Oslo, Norway. LNCS, Springer-Verlag. June, 2007. p. 160-163.
- E. Lupu, N. Dulay, M. Sloman, J.Sventek, S. Heeps, S. Strowes, K. Twidle, S.-L. Keoh, A. Schaeffer-Filho. AMUSE: Autonomic Management of Ubiquitous e-Health Systems. Concurrency and Computation: Practice and Experience, John Wiley and Sons, Inc., 2007 (To Appear). 
- [_URL_ Homepage] of Alberto E. Schaeffer-Filho
- Publications of Alberto E. Schaeffer-Filho, as indexed by DBLP