Full-text indexing in open source DBMS

Abstract

The main objective of this diploma thesis is to analyze and present the features of various open source SQL engines that perform text indexing for specific documents. We will explain the concept of text indexing and the fields used to enhance efficiency and reduce search time. Additionally, we will compare and enhance the performance of different Database Management Systems (DBMS) used for storing, processing, and searching large amounts of rapidly growing data. However, collecting and storing massive amounts of data alone does not address the issue of proper data organization and classification. To enable faster query processing and quick access to data records, indexing is necessary. It is crucial to carefully design indexes as they can oc cupy significant disk space. Hence, selecting the most suitable database management system and effectively managing the database are important considerations. For this diploma thesis, we will utilize MySQL, PostgreSQL, and SQLite as the database man agement systems. By using these systems, we aim to gain a comprehensive understanding of their construction and how text indexing can be implemented for specific documents. We will follow a consistent approach for all engines, covering fundamental text searching commands, text searching characteristics, auxiliary functions, and the creation of text indexing tables. Once we have gathered all the necessary data, we will conduct data visualization analysis to examine the response time of each system. This analysis will enable us to compare different functions not only within the same system but also across different SQL engines