ARCHITECTURAL MODEL OF THE METHOD FOR CONSTRUCTING A MICROSERVICE SOFTWARE PLATFORM IN THE EDGE/FOG ENVIRONMENT

Abstract

The paper proposes an architectural model of a microservice software platform for telecommunication networks in an Edge/Fog environment. In contrast to static service placement approaches, the proposed model integrates spatial grouping of user requests, network traffic identification, and multi-criteria Fog node selection into a single adaptive control loop. The platform state vector, the integral node evaluation function, and the adaptation trigger condition are formalized. A research software prototype was developed to reproduce the proposed architecture on a local central controller + fog agents testbed. Pilot testing confirmed the feasibility of the approach and showed a potential reduction in average response time and SLA violations under mixed-load scenarios.