Test tasks analysis of fault-tolerant software system multiversion formation

The relevance of the study is due to the modern requirements for the operational reliability of software systems for critical applications. The authors develop an approach based on modern information technology of highly reliable software system multiversion formation. The paper analyzes test tasks of fault-tolerant software system multiversion formation with the aid of ant colony algorithms including standard and modified algorithms. In this article, a software system is defined by a predefined set of software modules connected in a particular way and forming a transition graph with transition probabilities. Moreover, the execution of each module is multiversional, in other words, the module is comprised of several versions with each one characterized by the value of reliability and cost of execution. As a result, the set of versions, selected for execution in the module, determines its reliability and cost, and, owing to the presence of the program graph, we are able to calculate the reliability and cost of the entire software system. The conditions of the problem feature restrictions on the reliability and cost of the final solution. A predefined scheme of the software system was used in the analysis, taking into account the long-term mode of program functions implementation and the capacity to change program structure in the process of its implementation. It is shown that the employment of the modified algorithm provides an advantage not only in the quality of the objective function value, but also in the speed of improving this solution, which is especially important for practical purposes when implementing software systems in real time.

1. Kovalev I.V., Kovalev D.I., Brit A.A., Saramud M.V. Informacionnaja tehnologija dlja mul'tiversionnogo formirovanija otkazoustojchivyh programmnyh sistem. Sistemy upravleniya i informatsionnyye tekhnologii = Modern Innovations, Systems and Technologies. 2021;2(84):56–68. DOI: 10.36622/VSTU.2021.84.2.013 (In Russ.)

2. Buhovcev D.D. Primenenie modificirovannogo algoritma murav'inoj kolonii dlja reshenija zadachi kalendarnogo planirovanija raspredelennyh predprijatij. Sovremennyye innovatsii, sistemy i tekhnologii = Modern Innovations, Systems and Technologies. 2021;1(1):29–42. Available by: https://doi. org/10.47813/2782-2818-2021-1-1-29-42 (In Russ.)

3. Saramud M.V. K voprosu prognozirovanija vremeni narabotki na otkaz otkazoustojchivyh programmnyh kompleksov. Materialy XXIV Mezhdunarodnoj nauchno-prakticheskoj konferencii, posvjashhennoj pamjati general'nogo konstruktora raketno-kosmicheskih sistem akademika M. F. Reshetneva. Krasnoyarsk, 2020:453–454. (In Russ.)

4. Saramud M. V., Kovalev D. I. Sredstvo avtomatizirovannogo proektirovanija mul'tiversionnogo programmnogo kompleksa. Svidetel'stvo o registracii programmy dlja JeVM 2021610550, 15.01.2021. Zajavka № 2020667734, 28.12.2020. (In Russ.)

5. Ning J., Zhang C., Sun P., Feng Y. Comparative study of ant colony algorithms for multi-objective optimization. Information. 2019;10(1):11.

6. Jeona Young-Jae Kimb, Jae-Chul, Yunc Sang-Yun, Leed Kwang Y. Application of ant colony algorithm for network reconfiguration in distribution systems. IFAC Proceedings Volumes. 2003;36(20):773–778.

7. Borisenkov D.S. Preobrazovanie struktury programmy na jetape proektirovanija. Perspektivy razvitija informacionnyh tehnologij. 2014;22:15–20. (In Russ.)

8. Ponachugin A.V. Opredelenie nadjozhnosti programmnogo obespechenija v strukture sovremennoj informacionnoj sistemy. Kibernetika i programmirovanie. 2019;2:65–72. (In Russ.)

9. Zenjutkin N.V., Kovalev D.I., Tuev E.V, Tueva E.V. O sposobah formirovanija informacionnyh struktur dlja modelirovanija objektov, sred i processov. Modern Innovations, Systems and Technologies. 2021;1(1):10–22. Available by: https://doi.org/10.47813/2782-2818-2021-1-1-10-22 (In Russ.)

10. Gimarov V.V., Glushko S.I., Dli M.I. Primenenie algoritmov murav'inoj kolonii pri upravlenii slozhnymi proektami. Transportnoe delo Rossii = Transport business of Russia. 2012;4:107–109. (In Russ.)

11. Dorigo M., Stützle T. Ant colony optimization: Overview and recent advances. Handbook of Metaheuristics. International Series in Operations Research & Management Science. 2010;146:5–8. Available by: https://doi.org/10.1007/978-1-4419-1665-5_8.

12. Colorni A., Dorigo M., Maniezzo V. Distributed optimization by ant colonies. Proceedings of ECAL91 - European Conference on Artificial Life. 1991:134–142.