The rigorous application of static timing analysis requires a large and costly amount of detail knowledge on the hardware and software components of the system. Probabilistic Timing Analysis has potential for reducing the weight of that demand. In this paper, we present a sound measurement-based probabilistic timing analysis technique based on Extreme Value Theory. In all the experiments made as part of this work, the timing bounds determined by our technique were less than 15% pessimistic in comparison with the tightest possible bounds obtainable with any probabilistic timing analysis technique. As a point of interest to industrial users, our technique also requires a comparatively low number of measurement runs of the program under analysis, less than 650 runs were needed for the benchmarks presented in this paper.

Static timing analysis is the state-of-the-art practice of ascertaining the timing behavior of current-generation real-time embedded systems. The adoption of more complex hardware to respond to the increasing demand for computing power in next-generation systems exacerbates some of the limitations of static timing analysis. In particular, the effort of acquiring (1) detailed information on the hardware to develop an accurate model of its execution latency as well as (2) knowledge of the timing behavior of the program in the presence of varying hardware conditions, such as those dependent on the history of previously executed instructions. We call these problems the timing analysis walls. In this vision-statement article, we present probabilistic timing analysis , a novel approach to the analysis of the timing behavior of next-generation real-time embedded systems. We show how probabilistic timing analysis attacks the timing analysis walls; we then illustrate the mathematical foundations on which this method is based and the challenges we face in the effort of efficiently implementing it. We also present experimental evidence that shows how probabilistic timing analysis reduces the extent of knowledge about the execution platform required to produce probabilistically accurate WCET estimations.

The unabated increase in the complexity of the hardware and software components of modern embedded real-time systems has given momentum to a host of research in the use of probabilistic and statistical techniques for timing analysis. In the last few years, that front of investigation has yielded a body of scientific literature vast enough to warrant some comprehensive taxonomy of motivations, strategies of application, and directions of research. This survey addresses this very need, singling out the principal techniques in the state of the art of timing analysis that employ probabilistic reasoning at some level, building a taxonomy of them, discussing their relative merit and limitations, and the relations among them. In addition to offering a comprehensive foundation to savvy probabilistic timing analysis, this article also identifies the key challenges to be addressed to consolidate the scientific soundness and industrial viability of this emerging field.

Probabilistic Timing Analysis (PTA) in general and its measurement-based variant called MBPTA in particular can mitigate some of the problems that impair current worst-case execution time (WCET) analysis techniques. MBPTA computes tight WCET bounds expressed as probabilistic exceedance functions, without needing much information on the hardware and software internals of the system. Classic WCET analysis has information needs that may be costly and difficult to satisfy, and their omission increases pessimism. Previous work has shown that MBPTA does well with benchmark programs. Real-world applications however place more demanding requirements on timing analysis than simple benchmarks. It is interesting to see how PTA responds to them. This paper discusses the application of MBPTA to a real avionics system and presents lessons learned in that process.

Graphics Processing Unit (GPU) devices and their associated software programming languages and frameworks can deliver the computing performance required to facilitate the development of next-generation high-performance safety-critical systems such as autonomous driving systems. However, the integration of complex, parallel, and computationally demanding software functions with different safety-criticality levels on GPU devices with shared hardware resources contributes to several safety certification challenges. This survey categorizes and provides an overview of research contributions that address GPU devices’ random hardware failures, systematic failures, and independence of execution.

GPUs have been shown to cover the computing performance needs of autonomous driving (AD) systems. However, since the GPUs used for AD build on designs for the mainstream market, they may lack fundamental properties for correct operation under automotive's safety regulations. In this paper, we analyze some of the main challenges in hardware and software design to embrace GPUs as the reference computing solution for AD, with the emphasis in ISO 26262 functional safety requirements.