Let's get less optimistic in measurement-based timing analysis

Abstract

???This material is presented to ensure timely dissemination of scholarly and technical work. Copyright and all rights therein are retained by authors or by other copyright holders. All persons copying this information are expected to adhere to the terms and constraints invoked by each author's copyright. In most cases, these works may not be reposted without the explicit permission of the copyright holder." ???Copyright IEEE. Personal use of this material is permitted. However, permission to reprint/republish this material for advertising or promotional purposes or for creating new collective works for resale or redistribution to servers or lists, or to reuse any copyrighted component of this work in other works must be obtained from the IEEE.???Measurement-based timing analysis (MBTA) is a hybrid approach that combines execution time measurements with static program analysis techniques to obtain an estimate of the worst-case execution time (WCET) of a program. In order to minimize the chance that the WCET estimate is below the real WCET, the set of representative execution-time measurements has to be selected advisedly. We present an input data generation technique that uses a combination of model checking and genetic algorithms in order to heuristically optimize the set of measurements in terms of safety

Establishing Confidence and Understanding Uncertainty in Real-Time Systems

“…is was extended in [7] which combines the results produced with a genetic algorithm, which then aims to identify larger execution times. One drawback is that the tool analyses so ware that has been simpli ed to ensure each decision point relies on only a single variable.…”

“…Next a statistical analysis was performed across all 100 trials. Figure 3 presents another density plot, however this time across the set of 100 trials where the x-axis is calculated according to Equation (7). e x-axis represents the Coe cient of Variation, i.e.…”

Taco

“…Their research focuses on identifying efective coverage metrics to drive a model checking test suite generator. This was extended in [2] which combines the results produced with a genetic algorithm, which then aims to identify larger execution times. Drawbacks include the software being analysed has to be simpliied to ensure each decision point relies on only a single variable and the tool's use of model checking risks the tool's portability to larger, more complex functionality.…”

“…Overall, more than 95% of the 1800 test code items have been successfully analysed through the framework. Some items are not amenable to testport generation in the absence of some contextual 2 22 million test vectors were tested by the BCHLr heuristic on the i686 platform in less than 48 Hours, where 2000 iterations take an average of 1 Hour on target [14] because of data upload/download overheads. Figure 3: Test code items successfully analysed through TACO knowledge, e.g.…”

Context-Sensitive Measurement-Based Worst-Case Execution Time Estimation

“…That can be a structural coverage goal, e.g., a requirement like basic block, condition or decision coverage, or more sophisticated specifications [9], [10]. Alternatively, a coverage goal can be an optimization goals, e.g., maximization of locally observed execution times [11].…”