Development of a high availability method for configuration management software

Abstract

The article proposes its own method of providing high availability for configuration management software. The current state of the electronic resources management sphere was examined, the reasons for the use of automation tools were provided. The advantages of using configuration management software were analyzed, examples of using Infrastructure as Code and GitOps approaches to automate the deployment and scaling of electronic resources were given. The existing methods of ensuring high availability were analyzed. The development of our own method of ensuring high availability was carried out. The resulting method of providing high availability is based on the Raft consensus algorithm and the software system clustering approach and extends them with its own solutions. The algorithm of the proposed method was developed, the resulting flowchart of the algorithm and individual steps of its implementation were described in detail. The efficiency of the developed method was evaluated. An a priori ranking of a number of factors that evaluate the effectiveness of automatic recovery strategies and methods was conducted. The analysis of the results has shown that the proposed method implements the most important factors for experts, and in terms of RTO and RPO, the method can work on a par with existing popular disaster recovery strategies.