RESEARCH OF LIMIT TIME INDICES OF SOFTWARE ENVIRONMENTS OF REAL TIME OPERATING SYSTEMS
DOI:
https://doi.org/10.15802/stp2018/133384Keywords:
worst case execution time (WCET), real-time operating systems (OS RT), real-time tasks, hybrid method, static method, dynamic methodAbstract
Purpose. More than 20 years of the Worst Case Execution Time (WCET) studies have led to the development of many methods for its evaluation. So far, there are no definitive conclusions about usage of these methods. Therefore, the purpose of this paper is to determine the possibility of using a hybrid method for estimating WCET in real-time systems. Methodology. The approach for evaluating WCET for a hybrid method is to parse the input code in the C++ programming language and, after constructing the control flow graph, get the execution time of its base blocks. After finding the longest way of the graph, to estimate the time of execution of this way and get the WCET evaluation. To find the longest-running way, the reverse Dijkstra algorithm was chosen. After that, there was made a comparison of the limiting time estimations that were obtained by static and hybrid methods, as well as an analysis of the discrepancy between these results. Findings. Determining the worst execution time of programs is most important for "hard real-time" tasks. Underestimation of this indicator can lead to catastrophic consequences. An overestimation – to a significant overexpenditure of resources. Therefore, WCET was evaluated using static and dynamic methods. It was determined that the results obtained by the two methods correlate well. For the class of tasks that are under consideration, WCET execution time can be determined using a hybrid method. Adding an extra time of 15–20 % to the results, obtained by the hybrid method, allows getting a foolproof estimate of WCET in real-time systems. Originality. The hybrid method has been existing for some time, but the reliability of its use has not been sufficiently studied yet. The given study makes a step in determining the practical applicability of the hybrid method for time estimation of real-time tasks. Practical value. The results of this research allow us to conclude that a hybrid method could be used for obtaining WCET in "hard real-time" tasks. Furthermore, some unknown at this moment influence of the OS RT environment can be taken into account by adding extra time.
References
Balashov, V. V. (undated). Vstroennye informatsionno-upravlyayushchie sistemy realnogo vremeni. Lektsiya 3: Otsenka naikhudshego vremeni vypolneniya programm (WCET). Retrieved from http://lvk.cs.msu.su/~sveta/Балашов_3_Оценка_WCET.pdf (in Russian)
Kozyrev, V. P. (2016). Estimation of the execution time in real-time systems. Programming, 42(1), 39-50. (in Russian)
Betts, A. (2010). Hybrid Measurement-Based WCET Analysis using Instrumentation Point Graphs. PhD thesis. University of York, Heslington. (in English)
Corti, M., Brega, R., & Gross, T. (2001). Approximation of Worst-Case Execution Time for Preemptive Multitasking Systems. In Languages, Compilers, and Tools for Embedded Systems. Lecture Notes in Computer Science (pp. 178-198). Berlin. doi: 10.1007/3-540-45245-1_12 (in English)
Damschen, M., Bauer, L., & Henkel, J. (2016). Extending the WCET Problem to Optimize for Runtime-Reconfigurable Processors. ACM Transactions on Architecture and Code Optimization, 13(4), 1-24. doi: 10.1145/3014059 (in English)
Damschen, M., Bauer, L., & Henkel, J. (2017). Timing Analysis of Tasks on Runtime Reconfigurable Processors. IEEE Transactions on Very Large Scale Integration (VLSI) Systems, 25(1), 294-307. doi: 10.1109/TVLSI.2016.2572304 (in English)
Fellger, W., Gepperth, S., & Krause, F. (2011). Comparison of WCET Tools: Fachstudie Nr. 142. Stuttgart: Institute of Software Technology University of Stuttgart. (in English)
Lokuciejewski, P., & Marwedel, P. (2011). Worst-Case Execution Time Aware Compilation Techniques for Real-Time Systems. New York: Springer. doi: 10.1007/978-90-481-9929-7 (in English)
Puschner, P., & Nossal, R. (1998). Testing the results of static worst-case execution-time analysis. Proceedings of the IEEE Real-Time Systems Symposium, December 02-04, 134-143. doi: 10.1109/real.1998.739738 (in English)
Sewell, T., Kam, F., & Heiser, G. (2017). High-assurance timing analysis for a high-assurance real-time operating system. Real-Time Systems, 53(5), 812-853. doi: 10.1007/s11241-017-9286-3 (in English)
Stappert, F., & Altenbernd, P. (2000). Complete worst-case execution time analysis of straight-line hard real-time programs. Journal of Systems Architecture, 46(4), 335-339. doi: 10.1016/S1383-7621(99)00010-7 (in English)
Wilhelm, R., Engblom, J., & Ermedahl, A. (2008). The worst-case execution time problem – overview of methods and survey oftools. Transaction on Embedded Computing Systems, 7(3). doi:10.1145/1347375.1347389 (in English)
Downloads
Published
How to Cite
Issue
Section
License
Copyright and Licensing
This journal provides open access to all of its content.
As such, copyright for articles published in this journal is retained by the authors, under the terms of the Creative Commons Attribution 4.0 International License (CC BY 4.0). The CC BY license permits commercial and non-commercial reuse. Such access is associated with increased readership and increased citation of an author's work. For more information on this approach, see the Public Knowledge Project, the Directory of Open Access Journals, or the Budapest Open Access Initiative.
The CC BY 4.0 license allows users to copy, distribute and adapt the work in any way, provided that they properly point to the author. Therefore, the editorial board of the journal does not prevent from placing published materials in third-party repositories. In order to protect manuscripts from misappropriation by unscrupulous authors, reference should be made to the original version of the work.
