Performance Analysis Model for Fog Services under Multiple Resource Types

The rapid and widespread adoption of Internet of Thing-related services drives the advancement of Fog Computing, which is proposed to extend the Cloud Computing paradigm to run geo-distributed applications throughout the network. Fog services could be provisioned to tenants by applying container-based virtualization or system virtualization technology. Within a container/VM (virtual machine), an application can be configured and run. This paper aims to evaluate the performance of Fog services under different physical resource types, such as physical core and memory. For each type of physical resources, the resource amount demanded from different tenants is different and follows a general distribution. These differences make container/VM heterogeneous. We develop a novel analytical model to evaluate the performance of such heterogeneous containers/VMs deployed on the same fog node (a physical device) by applying the Continuous Time Markov Chain. Numerical results obtained from the proposed analytic model are verified through discrete-event simulations. The developed model and computing formulas could help the decision on how to schedule latency-sensitive IoT requests.