Article 3118

Title of the article



Bel'chenko Il'ya Vladimirovich, Postgraduate student, Kuban State Technological University (2 Moskovskaya street, Krasnodar, Russia),
D'yachenko Roman Aleksandrovich, Doctor of engineering sciences, associate professor, director of the Institute of Computer Systems and Information Security, Kuban State Technological University (2 Moskovskaya street, Krasnodar, Russia),

Index UDK





Backgrounds. The performance of the information system is one of the main indicators of its effectiveness. Most existing information systems and software complexes use relational databases to store information. Designing a database (DB) for an information system consists in the sequential development of a conceptual, logical and physical model. This design sequence has proven effective when the database requirements dictated by the subject area are strictly formalized, the load on the database server is predictable. But there are cases when the information systems working with the database are several. Each of them has its own information requirements, sometimes contradictory, which must be provided by the database in the shortest possible time. In such cases, the task of forecasting the load, allocating the most resource-intensive queries requires the use of methods of system analysis. The purpose of the work is the theoretical justification of the methodology for increasing the productivity of the information system through optimal data restructuring.
Materials and methods. Getting the optimal partition of the studied table structure into children begins with a system analysis of the methods of data storage by the database management system (DBMS). To formalize the domain, the parameters and sets that affect the processing speed of requests for reading information to the database table being researched are highlighted. An existing approach to storing rows of these tables as blocks on a disk drive is considered. The problem of optimization of the number of data blocks necessary for processing a group of requests for reading information is formulated. The objective function is proposed, structural limitations are introduced. A method for finding a suboptimal solution is proposed. The algorithm of the technique is described.
Conclusions. An approach is proposed for finding a suboptimal partition of the investigated table structure into its children. The proposed technique is especially relevant for database tables containing a large set of strings. The obtained results can be used in the design of domestic DBMS. Further research in this area is related to the development of techniques for searching for optimal partitioning of database tables without the need to introduce restrictions on the maximum possible number of child tables.

Key words

decision support system, optimization, data structures, databases, system analysis

Download PDF

1. Bogdanova A. V., D'yachenko R. A., Bel'chenko I. V. Politematicheskiy setevoy elektronnyy nauchnyy zhurnal Kubanskogo gosudarstvennogo agrarnogo universiteta [Polytechnical network electronic scientific journal of Kuban State Agrarian University].
2016, no. 117, pp. 873–885.
2. Embler S. V., Pramodkumar Dzh. Sadaladzh Refaktoring baz dannykh: evolyutsionnoe proektirovanie: per. s angl. [Database refactoring: evolutionary design: translated from English]. Moscow: Vil'yams, 2007, 672 p.
3. Tou D. Nastroyka SQL. Dlya professionalov [SQL configuration. For professionals]. Saint-Petersburg: Piter, 2004, 333 p.
4. Chigarkina E. I. Bazy dannykh: ucheb. posobie [Databases: scientific manuals]. Samara: Izd-vo SGAU, 2015, 208 p.
5. Atroshchenko V. A., Belchenko V. Ye., Belchenko I. V., Dyachenko R. A. International journal of pharmacy and technology. 2016, vol. 8, no. 2, pp. 14137–14149.
6. George J. Klir. Facets of Systems Science. New York, Springer, 1991, 664 p.


Дата создания: 13.06.2018 14:02
Дата обновления: 03.07.2018 16:16