Mathematical Models and Methods for Monitoring and Predicting the State of Globally Distributed Computing Systems
https://doi.org/10.31854/1813-324X-2021-7-3-73-78
Abstract
Monitoring events and predicting the behavior of a dynamic information system are becoming increasingly important due to the globalization of cloud services and a sharp increase in the volume of processed data. Well-known monitoring systems are used for the timely detection and prompt correction of the anomaly, which require new, more effective and proactive forecasting tools. At the CMG-2013 conference, a method for predicting memory leaks in Java applications was presented, which allows IT teams to automatically release resources by safely restarting services when a certain critical threshold value is reached. Article’s solution implements a simple linear mathematical model for describing the historical trend function. However, in practice, the degradation of memory and other computational resources may not occur gradually, but very quickly, depending on the workload, and therefore, solving the forecasting problem using linear methods is not effective enough.
References
1. Shchemelinin D., Mescheryakov S. Capacity Management of Java-based Business Applications Running on Virtualized Environment. Annual International Conference of the Computer Measurement Group, CMG 2013, 14–15 May 2013, London, UK. Computer Measurement Group Inc.; 2013. p.193–197.
2. Sharyy S.P. Computational Methods Course. Novosibirsk: Institute of Computational Technologies SB RAS Publ.; 2012. (in Russ.) URL: http://mathscinet.ru/files/SShary.pdf [Accessed 28th July 2021]
3. Kucherova K., Mescheryakov S., Shchemelinin D. Cloud monitoring – focusing on forecasting. Internet, Mobile, Performance and Capacity, Cloud and Technology Conference, IMPACT 2016, 07–10 November 2016, La Jolla, USA. Computer Measurement Group Inc.; 2016.
4. Gaysenok I.S., Zakrevskaya D.V. Assessment of the Quality of Forecasts. Proceedings of the Scientific and Practical Conferences on Business Education as a Tool for Sustainable Economic Development. Irkutsk: Irkutsk State University Publ.; 2012. (in Russ.)
5. Wolfram. Wolfram Mathematica Computations. URL: https://www.wolfram.com/mathematica [Accessed 28 July 2021]
6. Tikhonov E.Ye., Burdo A.I. On the Issue of Improving Automated Forecasting Systems. Proceedings of. the Interregional Conference on Student Science – Economics of Scientific and Technological Progress. Stavropol: North Caucasus Federal University Publ.; 2000. p.30–31. (in Russ.)
7. Sizov A.A. Models, Methods and Software for Decision Support Based on Forecasting Time Series with Variable Structure. PhD Thesis. Moscow: National Research University "Moscow Power Engineering Institute" Publ.; 2014. p. 141. (in Russ.)
8. Gildeh D. Monitoring Nightmares for DevOps. SlideShare. 2014. URL: https://www.slideshare.net/outlyer/david41050819 [Accessed 28 July 2021]
9. Bhandarkar M. Future of Data Intensive Applications. Proceedings of the 2nd ASE International Conference on Big Data Science and Computing. Stanford; 2014. URL: https://www.kdnuggets.com/2014/08/ase-conference-big-data-science-day1.html [Accessed 28 July 2021]
10. Kucherova K. Prediction of Cloud Computing Resources Based on the Open Source Monitoring System. Proc. of Telecom. Universities. 2020;6(3):100–106. (in Russ.) DOI:10.31854/1813-324X-2020-6-3-100-106
11. Logunov D.G. Application of Methods and Models for Forecasting Time Series for Assessing Cyclical Load in Cloud Systems. Proceedings of Scientific and Practical Conference with International Participation on Week of Science SPbPU, 18–23 November 2019, St. Petersburg, Russia. St. Petersburg: Politekh-Press Publ.; 2020. p.165–168. (in Russ.)
12. Mikhailov K.I. Analysis of Software Tools for Monitoring High-Load Cloud Information Systems. Proceedings of the XXIIIrd International Scientific and Practical Conference on System Analysis in Design and Management, SAEC-2019, 10–11 June 2019, St. Petersburg, Russia. Part 1. St. Petersburg: Politekh-Press Publ.; 2019. p.359–362. (in Russ.)
13. Mikhailov K.I. Monitor and Visualize Mixed Assets in the Cloud. Proceedings of Scientific and Practical Conference with International Participation on Week of Science SPbPU, 19–24 November 2018, St. Petersburg, Russia. Institute of Mechanical Engineering, Metallurgy and Transport. Part 2. St. Petersburg: Politekh-Press Publ.; 2018. p.234–236. (in Russ.)
14. Yakovlev K.A. Analytical Processing and Visualization of Monitoring Data of a Globally Distributed Information System. Proceedings of Scientific and Practical Conference with International Participation on Week of Science SPbPU, 13–19 November 2017, St. Petersburg, Russia. St. Petersburg: Peter the Great St.Petersburg Polytechnic University Publ.; 2017. p.203–205. (in Russ.)
15. Kucherova K.N. Monitoring and Forecasting Server Resources of Cloud Architecture Databases. Proceedings of Scientific and Practical Conference with International Participation on Week of Science SPbPU, 13–19 November 2017, St. Petersburg, Russia. St. Petersburg: Peter the Great St.Petersburg Polytechnic University Publ.; 2017. p.199–201. (in Russ.)
16. Kucherova K.N., Efimov V.V. Improving the Efficiency of Monitoring Cloud Infrastructure Based on Predictive Modeling. Proceedings of Scientific and Practical Conference with International Participation on Week of Science SPbPU, 14–19 November 2016, St. Petersburg, Russia. St. Petersburg: Peter the Great St.Petersburg Polytechnic University Publ.; 2016. p.117–119. (in Russ.)
Review
For citations:
Shchemelinin D. Mathematical Models and Methods for Monitoring and Predicting the State of Globally Distributed Computing Systems. Proceedings of Telecommunication Universities. 2021;7(3):73-78. (In Russ.) https://doi.org/10.31854/1813-324X-2021-7-3-73-78
Views:
1388
Email this article 