PDDM: um método de projeto de banco de dados aplicado à persistência poliglota

• Ano de defesa: 2019
• Autor(a) principal: Zdepski, Cristofer
• Tipo de documento: Dissertação
• Tipo de acesso: Acesso aberto
• Idioma: por
• Instituição de defesa: Universidade Tecnológica Federal do Paraná; Ponta Grossa; Brasil; Programa de Pós-Graduação em Ciência da Computação; UTFPR
• Palavras-chave em Português: Banco de dados não relacionais; Armazenamento de dados; Projeto lógico digital; Computação; Non-relational databases; Data Warehousing; Logic design; Computer science; CNPQ::CIENCIAS EXATAS E DA TERRA::CIENCIA DA COMPUTACAO; Engenharia/Tecnologia/Gestão
• Link de acesso: http://repositorio.utfpr.edu.br/jspui/handle/1/4718

In recent years, the growth of databases by Web 2.0 applications has revealed the limitations of the relational model related to scalability. This led to the emergence of NoSQL databases, with data storage models other than relational ones. These databases propose solutions to such limitations through horizontal scalability and partially compromise data consistency. The combination of multiple data models, called polyglot persistence, extends these solutions by providing resources for the implementation of complex systems that have components with distinct requirements that would not be possible by the use of only one data model in a satisfactory way. However, there are no consolidated methods for the NoSQL database design and neither methods for design systems that apply the polyglot persistence. This work proposes a database design method applied to systems that use polyglot persistence, combining different data models. This method can be applied to the relational model and aggregate-oriented NoSQL data models. The proposed method defines a set of sub-steps based on the existing concepts of database design. The goal is to define a formal process to assist in defining the data models to be used and to transform the conceptual design into a logical design. At the end, the method application are demonstrated in 3 test cases, in order to demonstrate its results and its applicability for later execution of the physical design of these databases.