Semantic Data Management in IT Service Performance Assurance

In today's dynamic and highly composed environments, IT service performance and dependability assurance require efficient reasoning about the performance and dependability effects of faults and the countermeasures to choose, using limited knowledge. Model- and observation-based qualitative error propagation analysis methods can be applied to this end; however, providing support for the human, as well as conceptually structured machine interpretation of sets of competing error propagation hypotheses is an open problem. This paper proposes the application of Formal Concept Analysis (FCA) for these tasks. A natural way to represent error propagation hypothesis sets as formal contexts is proposed, and the visual diagnostic exploration of formal context lattices is introduced. On this basis, potential applications of FCA in performance and dependability assurance activities are characterized.