[Back]

M. Zolda, R. Kirner, S. Bünte:
"Context-Sensitive Measurement-Based Worst-Case Execution Time Estimation";
Poster: 17th IEEE International Conference on Embedded and Real-Time Computing Systems and Applications (RTCSA'11), Toyama, Japan; 2011-08-28 - 2011-08-31; in: "Proc. 17th IEEE International Conference on Embedded and Real-Time Computing Systems and Applications (RTCSA'11)", (2011), ISBN: 978-1-4577-1118-3; 243 - 250.

The goal of measurement-based WCET estimation(MBWE) is to derive an estimate of the worst-case execution time (WCET) of a given piece of software on a particular target platform by executing the software on the target hardware and analyzing the obtained time-stamped execution traces. In this paper we introduce context-sensitive MBWE, an approach that can reduce pessimism by making use of state information that is exposed through individual control-flow decisions. We show how to extend the popular IPET method, to obtain tighter WCET estimates. We provide confirmative empirical results that demonstrate the effectiveness of our approach.