DATABASE DEPLOYMENT PATTERN FOR HIGH AVAILABILITY

Abstract

Ubiquitous digitalization and the development of big data technologies require a high level of availability (24/7) of information resources. Ensuring high availability of the database requires a significant amount of work on the design, testing and maintenance of several components and configurations. High availability can be achieved or maximized for the following deployment options: one database instance (with or without read replicas) and several main ones (with or without a coordinator). High availability ensures the smooth operation of the system by eliminating individual points of failure, while disaster recovery focuses on restoring normal operation after a failure or downtime. There are several deployment options because there is no universal solution. Choosing the best option depends on your specific use case and requirements. This article presents the main concepts of ensuring high availability and the most effective patterns for providing highly accessible database deployments. Purpose. definition of effective deployment patterns for database servers to ensure high availability. Methodology. The results of foreign and domestic scientific research are used in the work. The author uses theoretical research methods related to the search and analysis of information to identify connections and obtain unique conclusions. Results. The most effective deployment schemes for ensuring high availability and fault tolerance of databases are highlighted. Practical implications it is advisable to apply the results obtained in the field of DevOps development in order to optimize the process of developing and releasing applications by eliminating a known bottleneck: changes to the database code.

Keywords

• devops
• data management
• data platform
• database
• dbms
• high availability
• mysql
• mssql
• postgresql
• patterns