Rômulo OliveirapWCET Reliability Evaluation for Measurement-Based Probabilistic Timing Analysis
Universidade Federal de Santa Catarina Florianópolis, Brazil
CISTER, Porto, Portugal
The technique known as Measurement-Based Probabilistic Timing Analysis (MBPTA) promises to produce Worst-Case Execution Time (WCET) bounds for real-time systems’tasks based on the analysis of execution time measurements. It is based on Extreme Value Theory (EVT), a statistical framework designed to estimate the probability of extreme events. For that MBPTA requires the analysed tasks’ maximum observed execution times to adhere to an extreme value distribution, and allows determining execution time values expected to be exceeded only with arbitrarily small probabilities.
In this talk, we perform an empirical assessment on the reliability and tightness of the pWCET estimates yielded by the GEV and Gumbel models while applying MBPTA through the Block Maxima (BM) approach. We also consider pWCET estimates yielded by the GP and Exponential models when using the Peaks Over Threshold (POT) approach.
We do so by comparing the yielded estimates and their associated confidence intervals against the maximum values observed on large samples (e.g. of size 10^, of both real and synthetic nature, as the modelling sample size is increased.
Rômulo Silva de Oliveira received his PhD in Electrical Engineering from the Universidade Federal de Santa Catarina, Brazil, in 1997. He is currently professor in the Department of Automation and Systems of the Federal University of Santa Catarina (UFSC) and also serves on the Graduate Program in Engineering of Automation and Systems at UFSC. His main topics of interest are real-time systems, scheduling and operating systems. He is a co-author of more than 100 refereed publications in scientific conferences and journals.