Methods for multi-dimensional robustness optimization in complex embedded systems

Abstract

Design space exploration of embedded systems typically focuses on classical design goals such as cost, timing, buffer sizes, and power consumption. Robustness criteria, i.e. sensitivity of the system to variations of properties like execution and transmission delays, input data rates, CPU clock rates, etc., has found less attention despite its practical relevance.In this paper we introduce multi-dimensional robustness metrics, expressing the static and dynamic design robustness of a given system, the former assuming a fixed parameter configuration, and the latter including parameter adaptations as response to property variations. Additionally, we propose a metric measuring the robustness gain that can be achieved through system reconfigurability.Since determining multi-dimensional robustness is computationally expensive we introduce efficient exploration methods based on a stochastic sensitivity analysis technique capable of deriving upper and lower robustness bounds for a given system with low computational effort. We demonstrate the robustness optimization methods by means of a small but realistic case study.

Optimizing the Software Architecture for Extensibility in Hard Real-Time Distributed Systems

“…Also, a model of event-based activation for task and messages was assumed. In [9], [11], and [10], a generalized definition of extensibility on multiple dimensions (including changes in the execution times of tasks, as in our paper, but also period speedups and possibly other metrics) was presented. A randomized optimization procedure based on a genetic algorithm was used to solve the optimization problem.…”

Design synthesis and optimization for automotive embedded systems

“…In [10], task allocation and priority assignment are defined with the purpose of optimizing the extensibility with respect to changes in task computation times. In [26,25], a generalized definition of extensibility on multiple dimensions is presented, and a randomized optimization procedure based on a genetic algorithm is proposed to solve the optimization problem.…”