Dynamic allocation of microservices for virtual reality content delivery to provide quality of experience support in a fog computing architecture
Derian Alencar, Helder May Nunes da Silva Oliveira, Denis Lima do Rosário · 2023
Virtual Reality (VR) content is gaining popularity, demanding solutions for its efficient distribution over the Internet. Microservices present an ideal model for deploying services at different levels of a Fog computing architecture for managing traffic and provide Quality of Experience (QoE) guarantees to VR content. However, it is crucial to efficiently find the fog node to allocate the microservices, which directly impact the QoE of VR services. This paper proposes INFORMER, an integer linear programming model aiming to optimize the placement of caching services according to delay, migration time, and resource utilization rate. Based on the insights from INFORMER, the paper presents Fog4VR, a mechanism for the dynamic allocation of content based on a heterogeneous microservice architecture. Results obtained with INFORMER serve as a baseline to evaluate Fog4VR on different scenarios using a simulation environment. Results demonstrate the efficiency of Fog4VR compared to existing mechanisms in terms of cost, migration time, fairness index, and QoE.