Virtualization at the network edge: A performance comparison

Abstract

The rapid growth of the Internet of Things (IoT) is driving the evolution of Cloud technologies towards resource distribution across multi-clouds and the inclusion of various heterogeneous devices. IoT specific applications may require the deployment of gateways at the network edge to enable its interaction with physical sensors, pre-processing data from these sensors, and synchronizing it with the cloud. The orchestration, deployment, and maintenance of the software running on the gateways in large-scale deployments is known to be challenging. In this paper we analyze two virtualization techniques commonly used for deployment and management of applications at the cloud, but on the context of IoT and network edge. A study was conducted to evaluate the performance of the container-based approach compared to a hypervisor-based virtualization when running on devices typically used at the network edge. The study was performed through the execution of several synthetic benchmarks providing an insight in the performance overhead introduced by Docker containers (lightweight-virtualization) and KVM VMs (hypervisor-virtualization) running at network edge devices.