Workload correlations in multi-processor hard real-time systems

Modern embedded systems that are integrated as multi-processor system on chips, are often characterized by the complex behaviors and dependencies between system components. Different events that trigger such systems may cause different execution demands, depending on their event type as well as on the task they are processed by, leading to complex workload correlations. For example in data processing systems, the size of an event's payload data will typically determine its execution demand on most or all system components, leading to highly correlated workloads. Performance analysis of such systems is often difficult, and conventional analysis methods have no means to capture the possible existence of workload correlations. This leads to overly pessimistic performance analysis results, and thus to expensive system designs with considerable performance reserves. We propose an abstract model to characterize and capture workload correlations present in a system architecture, and we show how the captured additional system information can be incorporated into an existing framework for modular performance analysis of embedded systems.