A survey of NewSQL DBMSs focusing on taxonomy, comparison and open issues

Advancements in internet, cloud, and business intelligence technologies, as well as the emergence of big data, have caused massive traffic in online transaction processing (OLTP) systems.All these impose the needs for effective and efficient data storage and processing which is not available in conventional Relational Database Management Systems (RDBMSs). In the year2000, a new generation of database management systems (DBMSs)called Not Only Structured Query Language (NoSQL) has emerged to address the scalability of OLTP workloads in a way that was impossible for conventional relational database management systems(RDBMSs). Despite adequately addressing issues linked to scalability and producing better read-write performance than RDBMS,NoSQL DBMSs appear to lack in providing definite assurance of data consistency. Hence, a newer DBMS alternative called NewSQLhas emerged. NewSQL DBMS has combined the advantages of both conventional RDBMS and NoSQL. They have the scalable performance of NoSQL DBMSs and the data consistency of traditional RDBMS. To date, there are tens of existing NewSQL-DBMSs,so it is difficult to understand the difference between them, andit is also quite challenging to decide the best solution for a specific task. Hence, this paper presents a comprehensive review concerning NewSQL-DBMSs by emphasizing the following purposes:(1) providing researchers and practitioners guidance that may assist in selecting the most appropriate NewSQL-DBMS, (2) classifying NewSQL-DBMSs based on internal implantation and number of tiers, (3) providing comparisons and analyses of query capabilities,technical characteristics, and security attributes of the prominent NewSQL-DBMSs, and finally, (4) identifying issues and challenges raised with the emergence of NewSQL-DBMSs. In conclusion,NewSQL DBMSs can offers solutions for fault tolerance, horizontal scalability features.

Actions (login required)

View Item