<div class='slidealt'>Experience kvm <a title='virtualization for embedded heterogeneous arm core platforms' href='/en/products'>virtualization extensions</a></div> <div class='slidealt'>Benefit from custom <a title='kvm on arm services full virtualization' href='/en/services'>virtualization services</a></div> <div class='slidealt'>KVM on ARMv7 and ARMv8 <a title='kvm-on-arm open source smu extensions' href='/en/solutions/guides/vfio-on-arm/'>IOMMU full virtualization</a></div> <div class='slidealt'>Virtualization research projects <a title='ARM multicore kvm open source' href='/en/research'>in cloud and embedded systems</a></div> <div class='slidealt'>Virtualization solutions for heterogeneous <a title='ARMv7-ARMv8 virtualization open source solutions' href='/en/solutions'>ARM multicore systems</a></div>

Virtual Open Systems Scientific Publications

Virtualized Infrastructure Managers for edge computing: OpenVIM and OpenStack comparison

Virtualized Infrastructure Managers for edge computing: OpenVIM and OpenStack comparison

Event

The IEEE International Symposium on Broadband Multimedia Systems and Broadcasting 2018 (BMSB 2018), Valencia, Spain.

Keywords

Virtual Infrastructure Manager, OpenStack, OpenVIM, Edge Computing, CloudBench, Open Source, Benchmark, Performance.

Authors

Teodora Sechkova, Michele Paolino, Daniel Raho.

Abstract

The evolution of the centralized cloud computing architecture towards the edge of the network, driven by virtualization technologies such as Software-Defined Networking (SDN), Network Functions Virtualization (NFV) and Multi-access Edge Computing (MEC), brings new opportunities to the multimedia industry. Offloading computing power to the edge, local caching, minimized latency and flexibility in the deployment of services are only some of the benefits that the multimedia can gain from utilizing the edge computing capabilities. To achieve such progress the potential elements of an edge computing framework have to be studied and evaluated. In this work, we focus on the component managing the edge resources which is defined as the Virtual Infrastructure Manager (VIM) in the MEC reference architecture. We analyze the time overhead which virtual machines (VMs) provisioning brings into the system. This is done by a comparison of two popular open-source VIM solutions, OpenStack and OpenVIM. We discuss and compare the logical architectures of the two VIMs and present the results of their performance evaluation. We extended the open-source cloud benchmarking framework CloudBench by adding a new cloud adapter for OpenVIM and used it to gather the cloud management metrics.

Introduction

The recent convergence of key virtualization technologies such as Software-Defined Networking (SDN), Network Functions Virtualization (NFV), Multi-access Edge Computing (MEC) and Fifth Generation (5G) wireless systems, has pushed the traditional centralized cloud computing architectures to the edge of the network, in close proximity to the end devices. Multimedia applications are among the most beneficial from the cloud evolution which brings minimized latency, offloading network traffic and computing power, local caching, flexibility and added security. However, the edge computing paradigm will stay on only if it meets the newly opened challenges including scalability, usage of heterogeneous resource-constrained devices, limited network throughput etc.

The first step towards a successful application of edge computing principles is the evaluation of the different components involved in one such framework. In this work we focus on the management of an edge computing infrastructure which is performed by Virtualized Infrastructure Managers (VIMs) as defined by the European Telecommunications Standards Institute (ETSI). In particular we analyze the time overhead which virtual machines (VMs) provisioning brings into the system. We do a comparison of two widespread open-source solutions - OpenStack and OpenVIM by extending and utilizing the capabilities of an open-source benchmarking framework called CloudBench.

Contributed slides presentation

The slides presented at this conference are made publicly available.

Acknowledgement

This work has received funding from the European Union's Horizon 2020 research and innovation program, 5GCity, under grant agreement No 761508. The work presented in this paper reflects only authors' view and the European Commission is not responsible for any use that may be made of the information it contains.

Access the full content of this publication

Login or register to access full information