MODELS AND METHODS OF PROACTIVE MONITORING OF ITSYSTEMS

The high cost of organizations' IT systems and the significant losses from unplanned downtime caused by failures in such systems, make the search for new approaches to monitoring urgent. The most advanced is the proactive approach to monitoring, which is aimed not only at regular monitoring of the status of monitoring objects and responding to incidents, but also in forecasting possible emergencies at an early stage. This article is aimed at researching models and methods for proactive monitoring of IT systems. The main task of proactive monitoring is reduced to the task of forecasting time series taking into account external factors. To disclose the process under investigation, the classification of forecasting models of time series is considered and an overview of modern works devoted to models and methods for predicting the health of various components of the IT infrastructure is reviewed. The analysis showed that there is no generalized model that allows to solve any given task of forecasting the state of IT systems. The process of implementing the model and the corresponding method should be based on the capabilities of the selected classes of models and the requirements for solving the problem. During the review, a number of proposals are formulated that will improve the effectiveness of proactive monitoring and the quality of the forecast model. Conducting a comprehensive monitoring of the components of the IT system allows you to analyze the root causes that lead to an inoperative state of the system. For the correct detection of threshold values of the operable state of objects, it is necessary to use "dynamic thresholds". Forecasting the state of monitoring objects a few steps forward is an urgent issue for a distributed IT system.

1. Ardentov A. A., Moskovskiy A. A., Pervin A. Yu., Stotskiy M. V. Algoritmy prognozirovaniya apparatnykh sboev v sisteme monitoringa klasternykh ustanovok // XII nauchno-prakticheskaya konferentsiya universiteta goroda Pereslavlya. – 2008. No. 6 – pp. 84-95.

2. Boks Dzh., Analiz vremennykh ryadov, prognoz i upravlenie. / Dzh. Boks. G.M.Dzhenkins. – M.: Mir, 1974. – 406 p.

3. Petrakov V. A., Bogachev D. N. Primenenie neyronnykh setey v monitoringe vychislitel'nykh tsentrov // Izvestiya Yuzhnogo federal'nogo universiteta. Tekhnicheskie nauki. – 2009. No. 2. pp. 82-87.

4. Saenko I. B., Skorik F. A., Kotenko I. V. Monitoring i prognozirovanie sostoyaniya komp'yuternykh setey na osnove primeneniya gibridnykh neyronnykh setey //Izvestiya vysshikh uchebnykh zavedeniy. Priborostroenie. – 2016. – Vol. 59. – No. 10.

5. Rolik A.I., Timofeeva Yu.S., Turskiy N.I. Upravlenie ustraneniem neispravnostey v IT-sistemakh // Vestnik NTUU «KPI». Informatika, upravlenie i vychislitel'naya tekhnika. – 2008. No. 49. pp. 95-108

6. Tikhonov, E.E. Prognozirovanie v usloviyakh rynka. / E.E. Tikhonov. – Nevinnomyssk, 2006. – 221 p.

7. Trukhachev A.A., Ivkina E.A. Primenenie metodiki prognozirovaniya masshtabiruemosti dlya postroeniya sistem vysokoy dostupnosti na osnove SUBD Oracle // Spetstekhnika i svyaz'. 2011. No.6.

8. Fedorov E. E., Al'-Ababnekh Kh. A., Al'rababa Kh. Metod prognoza proizvoditel'nosti servernykh komp'yuternykh sistem //Naukovі pratsі Donets'kogo natsіonal'nogo tekhnіchnogo unіversitetu. Serіya: Іnformatika, kіbernetika ta obchislyuval'na tekhnіka. – 2015. – No. 1. – pp. 52-58.

9. Chuchueva I. A. Model' prognozirovaniya vremennykh ryadov po vyborke maksimal'nogo podobiya //Moskva. – 2012.

10. Shelupanov A.A., Iskhakov S.Yu., Timchenko S.V. Prognozirovanie v sisteme monitoringa lokal'nykh setey // Doklady Tomskogo gosudarstvennogo universiteta sistem upravleniya i radioelektroniki– 2012. No. 1-2 – pp. 100-103.

11. Alexander La rosa. Root Cause Analysis and Monitoring Tools: A Perfect Match. [Elektronnyy resurs] – Rezhim dostupa: URL: https://blog.pandorafms.org/root-cause-analysis/ (Data obrashcheniya: 24.01.2018)

12. Kothamasu R., Huang S. H., VerDuin W. H. System health monitoring and prognostics—a review of current paradigms and practices //The International Journal of Advanced Manufacturing Technology. – 2006. – Vol. 28. – No. 9- 10. – pp. 1012-1024.

13. .Rodd S. F., Kulkarni U. P. Adaptive tuning algorithm for performance tuning of database management system //arXiv preprint arXiv:1005.0972. – 2010.

14. Sahoo R. K. et al. Critical event prediction for proactive management in large-scale computer clusters //Proceedings of the ninth ACM SIGKDD international conference on Knowledge discovery and data mining. – ACM, 2003. – С. 426-435.