Prediction of Run-Time Resource Consumption in Multitask Component-Based Software Systems

Abstract. Embedded systems must be cost-effective. This imposes strict requirements on the resource consumption of their applications. It is therefore desirable to be able to determine the resource consumption of applications as early as possible in its development. Only then, a designer is able to guarantee that an application will fit on a target device. In this paper we will present a method for predicting run-time resource resource consumption in multi-task component based systems based on a design of an application. In [5] we describe a scenario based resource prediction technique and show that it can be applied to non-pre-emptive non-processing resources, like memory. In this paper we extend this tech-nique, which enables us to handle pre-emptive processing resources and their scheduling policies. Examples of these class of resources are CPU and network. For component based software engineering the challenge is to express re-source consumption characteristics per component, and to combine them to do predictions over compositions of components. To this end, we pro-pose a model and tools, for combining individual resource estimations of components. These composed resource estimations are then used in scenarios (which model run-time behavior) to predict resource consump-tion.