Using Architectural Runtime Verification for Offline Data Analysis

Abstract

Analyzing runtime behavior as part of debugging complex component-based systems used in the vehicle industry is an important aspect of the integration process. It is a laborious task that involves many manual steps. One reason for this is that, as of today, the analysis is usually not performed on the architecture level, where the system has initially been designed. Instead, it relies on source code debugging or visualizing signals and events. With an ever-growing complexity of such systems, it becomes increasingly difficult to find errors that manifest at integration level, i.e., when the components interact with each other in a complex environment. Architectural Runtime Verification (ARV) is an approach specifically designed for the integrator—a generic way to analyze system behavior on architecture level using the principles of Runtime Verification. This paper draws on our initial publication. It provides further details and an evaluation of the ideas using a database hosted in the cloud.