A Framework for Optimization, Service Placement, and Runtime Operation in the Fog

Abstract

Fog computing provides a paradigm for executing Internet of Things services. Enabling the coordinated cooperation among computational, storage, and networking resources in the fog can be challenging due to the volatility of resources. For this reason, we design an architecture and implement a representative framework called FogFrame that defines the necessary communication mechanisms for instantiating and maintaining service execution in the fog. To evaluate our approach, we conduct a series of experiments that show how service placement, deployment, and execution is performed by the framework, and how the framework operates at runtime, i.e., adapts to changes in the available resources, balances the workload and recovers from resource failures and overloads.