The emergence of pervasive computing in our everyday life supposes that the data necessary to the operation of the majority of our essential services in various fields of life will be managed by these systems. Thus, their dependability became a major concern. But, dependability issues have not been well explored so far in pervasive computing research. Pervasive environments are highly complex, heterogeneous and geographically dispersed. As a result, current means and facets of dependability do not address the needs of these systems. A solution to achieve this goal should be to adopt a dependability approach based on survivability in pervasive environments. But, the survivability suffers from a remarkable lack of suitable and mature methods for using it in practice. In this paper, we focus on achieving survivability in pervasive environments. First, we introduce a formal survivability model based on a rigorous definition of the concept of acceptable service and a method for calculating the degree of survivability of the system. Then, we present the basis for a new approach to adapt the system in adverse operation environment to comply with its survivability specification. To fix ideas, a case study in pervasive healthcare is presented.
展开▼
