Modeling Hierarchical Event Streams in System Level Performance Analysis

Abstract

The growing complexity of embedded real-time systems requires elaborate methods in their design process. While simulation does not sufficiently cover corner cases, analytic methods for system level performance analysis have been established in the past. Their efficiency in computing hard bounds on buffer sizes, end-to-end delays or throughput has proven their usefulness. One of the major drawbacks of these methods are the limited system configurations that can be analyzed with high accuracy. In this report we extend existing methods for analyzing heterogeneous multiprocessor systems such that different types of data streams can be composed to a higher level event stream with multiple hierarchies. Furthermore, we describe how such hierarchical event streams can be decomposed into their sub-streams. Additionally, we show how these new methodologies are used to model the distributed heterogeneous communication system of the COMBEST case study provided by EADS.