DIFFERENTIATED CAPACITY EXTENSION METHOD FOR SYSTEM OF DATA STORAGE WITH MULTILEVEL STRUCTURE

Tatarnikova T.M., Poymanova E.D. Differentiated capacity extension method for system of data storage with multilevel structure. Scientific and Technical Journal of Information Technologies, Mechanics and Optics, 2020, vol. 20, no. 1, pp. 66–73 (in Russian). doi: 10.17586/2226-1494-2020-20-1-66-73

Subject of Research. The paper presents a method for differentiated capacity extension of the data warehouse. The method is built on a predictive model of time series with an estimate of volume for the traffic storage. The effect of the incoming data stream structure on the choice of the prediction model is considered. Methods. The storage system is presented in the form of a matrix specifying the number of storage levels and the number of carriers/volumes at each level. The matrix elements are metadata of the recorded files that are stored on the corresponding carriers/volumes of multilevel data storage system. The matrix visualizes the data storage state in the form of patterns. Patterning is performed by systematic slices of matrix values. Periodic analysis of the data warehouse state patterns gives the possibility to evaluate the time to reach the maximum value of the carrier capacity. The predictive model, which is the basis of the method for data warehouse differentiated capacity extension, takes into account the structure of the incoming data stream. In the presence of a self-similar structure of traffic for storage, a predictive model of auto-regression and an integrated moving average is implemented. For traffic without a self-similar structure, a general linear predictive model of the time series at known past values is implemented. The prediction model is applied separately for each storage carrier/volume. Main Results. Structure features of the traffic arriving for storage are given. Self-similarity properties are verified on the example of LTE-traffic, demonstrating the presence of “heavy-tailed” distributions. The prediction results for volume of traffic arriving for storage are obtained by the autoregressive model and the integrated moving average. The predictive and real values of the traffic volume are given, as well as the prediction error value. A technique for differentiated capacity extension of the data storage system is developed, which establishes a sequence of steps for analysis of patterns and the structure of traffic arriving for storage. Practical Relevance. The method for differentiated capacity extension of the data storage takes into account the multilevel organization of storage and the structure of the incoming data stream, which provides organizing a differentiated capacity extension in accordance with the characteristics of the files and ensuring the requirements for guaranteed storage time.

1. Proskuryakov N.E., Anufrieva A.Y. Analysis and prospects of modern systems of storage of figures. News of the Tula state university. Technical sciences, 2013, no. 3, pp. 368–377. (in Russian)
2. Information Storage and Management. 2nd ed. New Jersey, John Wiley & Sons Inc., 2016, 544 p.
3. Farley M. Building Storage Networks. 2nd ed. Osborne, McGraw- Hall, 2001, 576 p.
4. Leonov V. Google Docs, Windows Live, and other cloud technologies.
Moscow, EKSMO Publ., 2012, 304 p. (in Russian)
5. Bogatyrev V.A., Bogatyrev S.V., Bogatyrev A.V. Reliability clusters computing systems with the duplicated communications of servers and storage devices. Information Technology, 2013, no. 2, pp. 27–32. (in Russian)
6. Mesnier M., Ganger G., Riedel E. Object-based storage. IEEE Communications Magazine, 2003, vol. 41 no. 8, pp. 84–90. doi: 10.1109/MCOM.2003.1222722

8. Burmistrov V.D., Zakovryashin E.M. Creating a data warehouse for a distributed system. Molodoi Uchenyi, 2016, no. 12, pp. 143–147. (in Russian)
9. Buyya R., Broberg J., Goscinski A.M. Cloud Computing: Principles and Paradigms. New Jersey, John Wiley & Sons Inc., 2011, 637 p.
10. Sovetov Ya.B., Tatarnikova T.M., Poymanova E.D. Organization of multi-level data storage. Informatsionno-Upravliaiushchie Sistemy, 2019, no. 2, pp. 68–75. (in Russian). doi: 10.31799/1684-8853-2019-2-68-75
11. Kish L.B., Granqvist C.G. Does information have mass? Proceedings of the IEEE, 2013, vol. 101, no. 9, pp. 1895–1899. doi: 10.1109/JPROC.2013.2273720
12. Morville P., Callender J. Search Patterns: Design for Discovery. O’Reilly Media, 2010, 192 p.
13. Stacey M., Salvatore J., Jorgensen A. Visual Intelligence: Microsoft Tools and Techniques for Visualizing Data. New Jersey, John Wiley & Sons Inc., 2013, 432 p.
14. Poymanova E.D., Tatarnikova T.M. Models and methods for studying network traffic. 2018 Wave Electronics and its Application in Information and Telecommunication Systems, WECONF 2018, 2018, pp. 8604470. doi: 10.1109/WECONF.2018.8604470
 

This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License