METHOD OF APPLYING MACHINE LEARNING TO ENHANCE THE EFFICIENCY OF DEVOPS PROCESSES

Abstract

Context. The speed and quality of software development determine the competitiveness of products, therefore integrating DevOps and machine learning through MLOps approaches opens new opportunities for optimization. The implementation of MLOps promises enhancements in automation, efficiency, and the quality of development, however, realizing these promises requires the development of practical tools and methods.

Objective. The goal of this study is to develop and evaluate a prototype framework based on MLOps to optimize DevOps processes. This prototype is designed to demonstrate how the integration of machine learning can enhance key aspects of DevOps, such as: testing automation, system monitoring, and deployment processes.

Method. The research methodology includes the development of a prototype that involves the selection and adaptation of machine learning tools for integration into DevOps. The prototype was experimentally tested on several projects to assess its impact on development speed and product quality.

Results. The results confirm that the application of the MLOps framework prototype contributes to significant improvements in the automation of DevOps processes, especially in the areas of testing and monitoring. This, in turn, leads to a reduction in development time and an increase in the quality of the final product. The study demonstrates that implementing an MLOps-based framework prototype into DevOps processes can effectively enhance their productivity and quality.

Conclusions. The conclusions from this study provide a valuable foundation for further development of MLOps tools and their application in the software development industry.