Resilience and load balancing in Fog networks: A Multi-Criteria Decision Analysis approach

The advent of Cloud Computing enabled the proliferation IoT applications for smart environments. However, distance these resources makes them unsuitable delay-sensitive applications. Hence, Fog has emerged to provide such capabilities in proximity end devices through distributed resources. These limited can collaborate serve application workflows using concept stateless micro service replicas, which provides resiliency and maintains availability face failures. Load balancing supports this collaboration by optimally assigning workloads appropriate services, i.e., distributing load among nodes fairly utilize compute network minimize execution delays. In paper, we propose ELECTRE, a Multi-Criteria Decision Analysis (MCDA) approach, efficiently balance We considered multiple objectives make selection decisions, including information. evaluate our approach realistic unbalanced topological setup with heterogeneous workload requirements. To best knowledge, is first time ELECTRE-based methods are used Through simulations, compared performance proposed traditional baseline that commonly practice, namely random, Round-Robin, nearest node, fastest algorithms. terms overall system performance, outperforms up 67% improvement.