SECURITY PROBLEMS OF UNIVERSAL DATA MANAGEMENT SYSTEMS

Abstract

The article deals with the security of universal data management systems. The analysis and classification of modern data management systems by different criteria has been made. Based on the analysis of the literature and the experience of creating corporate systems, two approaches to the organization of universal data management systems have been identified: the use of multimodel systems and integrated data management platforms. Based on the analysis of threats and data protection tools for database management systems SQL, NoSQL, NewSQL, Data Warehouse, Data Lake and data clouds, the main approaches to data protection of each product category have been identified. The current trends in the development of data management technologies and data security have been identified. The development of NoSQL, NewSQL systems and the exchange of functionalities between them has led to the development of systems, which have functions of many classes. The problems of data protection for multimodel database management systems and for integrated data platforms have been identified and ways to overcome the identified problems have been suggested. For a universal data management platform, it is not enough to combine security features of different types of DBMS but new approaches are needed. The Data Centric Security approach is suitable for integrated environments; it is focused on protecting critical data at all stages of their processing - from collection and transmission to analysis and deployment in data warehouses. The organization of access to data through logical data marts using semantic technologies, ontological data models provides the transformation of a set of different types of data into a single array by "data virtualization". The article has substantiated the relevance and feasibility of the use of cognitive technologies and artificial intelligence in the field of information security, which opened new opportunities for the creation of automated, "smart" security tools for data management systems. Such systems have the ability to self-analyse and configure. The use of machine learning technology allows to identify weaknesses in the database security system. The combination of intelligent security and management solutions with database technologies will allow developers to respond quickly to new challenges in the protection of integrated data management systems of various types.