Mobility-aware Service Function Chaining Orchestration for Multi-user Augmented Reality
Hugo Santos, Bruno Martins, Denis Lima do Rosário, Eduardo Cerqueira, Torsten Ingo Braun · 2023
Multi-User Augmented Reality (MUAR) is gaining popularity and enabling interactions collaboratively with mobile users in the virtual 3D world. In this way, decomposing MUAR into elements with specialized ordered Service Functions (SFs) to form a SF chain with decoupled source dynamically allows the SF distribution into multiple edge computing servers and executes SFs of MUAR services in parallel while improving its quality level. However, orchestrating distributed SFs at the network edges and at multiple mobile clients while minimizing latency and meeting resource-hungry requirements is still a research challenge. This paper proposes a mobility-aware SF chaining orchestration scheme for MUAR services called MSF. MSF improves the usage of processing, storage, and networking resources and reduces the latency of MUAR services by optimizing SF chaining in real-time and reinstantiating services into optimal routes. The results show that MSF outperforms state-of-the-art approaches regarding MUAR session acceptance ratio, CPU and bandwidth utilization, and latency in different mobile scenarios.