INFORMATION TECHNOLOGY FOR OPTIMAL CONFIGURATION SELECTION IN AUTOMATED TESTING OF MULTICOMPONENT INFORMATION SYSTEMS: REVIEW AND PROPOSAL

Abstract

The article analyzes modern approaches to selecting optimal configurations for automated testing of multi- component information systems (IS), which are the basis of the functionality of digital platforms. Key concepts are defined: configuration as a combination of parameters (software versions, databases, browsers), optimal configuration as a minimumset of combinations to cover critical scenarios under resource constraints, multi-component IS as a set of interconnected components (frontend, backend, API, databases).The objective of the study is to evaluate current methods and create an integrated approach that combines combinatorialmethods, genetic algorithms, and CI/CD to automate configuration selection in real time.Methodology includes a systematic review of the literature over the last 5 years, comparative analysis using weighting factors (number of tests, scenario coverage, adaptability, integration with CI/CD, resources), mathematical modeling, and testing on the example of cloud platforms and e-commerce systems.The scientific novelty lies in developing an integrated approach that reduces the number of tests to 5–10% of the full set (for example, from 243 to 12–24 configurations for a system with 5 parameters), while ensuring 90–95% coverage of criticalscenarios and high adaptability to component changes. The uniqueness of the approach is integration with CI/CD processesand the use of weight analysis to select optimal configurations.Conclusions. The proposed approach allows optimizing testing in complex IS environments, combining the accuracy ofcombinatorial methods, the efficiency of genetic algorithms, and CI/CD automation. Prospects for further research include theuse of AI for defect prediction and automatic analysis of test results.