ANALYSIS OF METHODS FOR REPRESENTING TREE-LIKE STRUCTURES IN A RELATIONAL DATA MODEL

Abstract

"As you know, the data must be stored in such a way that their sampling takes place with minimal expenditure of time and resources. In most cases, it is convenient and expedient to use the relational data model, which has found its application in relational databases, for data storage and processing. The advantages of this data representation are the fact that the entities are represented in the form of tables, connected to each other by various relationships: ""one-to-one"", ""one-to-many"" and ""many-to-many"". Thus, we have a single system of interconnected entities that represent some subject area. Each entity has a set of fields that have specific data types: numeric, text, date/time, and others. In addition, tables have a key field that uniquely identifies an entry in the table. It is this field that plays an important role in building relationships between tables. This field includes records of related tables, through a foreign key, that they have in their list of fields. With the correct design of such a system, it should correspond to the third normal form. Presenting tree-like structures in the relational data model is a very relevant task today. A typical example of such data is categories, which can be quite a lot with a high level of nesting in each other. The article considers and analyzes the most popular methods of representing tree structures in the relational data model. A typical example of such data is categories, which can be quite a lot with a high level of nesting in each other. Such methods include Adjacency List and Nested Sets. Advantages and disadvantages of these methods were identified. Quantitative indicators of time for data sampling, which are presented in the database by the considered methods, were obtained. On the basis of these indicators, an analysis of the expediency of data presentation by these methods was carried out, depending on some characteristics of nestedness and data volumes. The advantages and disadvantages of the considered methods for representing tree-like structures in the relational data model are revealed. "