When Relational-Based Applications Go to NoSQL Databases: A Survey

Abstract

Several data-centric applications today produce and manipulate a large volume of data, the so-called Big Data. Traditional databases, in particular, relational databases, are not suitable for Big Data management. As a consequence, some approaches that allow the definition and manipulation of large relational data sets stored in NoSQL databases through an SQL interface have been proposed, focusing on scalability and availability. This paper presents a comparative analysis of these approaches based on an architectural classification that organizes them according to their system architectures. Our motivation is that wrapping is a relevant strategy for relational-based applications that intend to move relational data to NoSQL databases (usually maintained in the cloud). We also claim that this research area has some open issues, given that most approaches deal with only a subset of SQL operations or give support to specific target NoSQL databases. Our intention with this survey is, therefore, to contribute to the state-of-art in this research area and also provide a basis for choosing or even designing a relational-to-NoSQL data wrapping solution.

A Comprehensive Spark-Based Layer for Converting Relational Databases to NoSQL

“…The dataset used in this experiment was small. SQLtoKeyNoSQL [7,10,18] uses a canonical model for transformation and migration to key-value, document, and column databases. This supports merge and hash join.…”

Um Estudo sobre Modelagem Poliglota de Dados

“…Com o surgimento dos Bancos de Dados (BDs) NoSQL houve uma melhora substancial em termos de escalabilidade para gestão de dados complexos e heterogêneos. Neste cenário, algumas pesquisas buscam apresentar abordagens de migrac ¸ão de BDs SQL para NoSQL a fim de mitigar as deficiências encontradas nos SGBDRs [Schreiner et al 2019]. As vantagens em relac ¸ão a escalabilidade horizontal e a flexibilidade de estrutura que os sistemas de armazenamento NoSQL possuem os tornam opc ¸ões muito interessantes no momento de projetar a infraestrutura de dados de um sistema de informac ¸ão.…”

Complex SQL-NoSQL Query Translation for Data Lake Management

“…A data lake is a system for storing structured, semi-structured, unstructured, and binary data of any type in a format that exhibits 4V Big Data is a large fact function inclusive of volume, veracity, variety, and veracity (Khine and Wang, 2018). NoSQL databases can store different types of data, but they cannot fully support ACID (Automation, Integrity, Isolation, and Durability) features, i.e., the trigger functions in multi-transaction management (Lotfy et al, 2016), because a NoSQL database uses a non-relational database system (Schreiner et al, 2019;Li and Gu, 2019).…”