Análise do Desempenho da Virtualização Leve para Ambientes com Edge Computing baseada em NFV

• Leon Valentim Porto Trindade UFRJ
• Luís Henrique M. K. Costa UFRJ

DOI: https://doi.org/10.5753/sbrc.2018.2464

Abstract

The Network Functions Virtualization (NFV) paradigm is changing the way network services are provided, allowing the same degree of versatility and agility as ever seen in cloud computing. However, it is possible to implement as virtual functions through traditional virtualization in hypervisors or in lightweight virtualization, constituting a compromise between flexibility and performance. Thus, this work compares the performance in environments that use virtualization technology with hypervisor to those that uses a container, for the purpose of verify the best for operational environments allied to the concept of Edge Computing in NFV. However, it is intended to gauge the ability for the technologies to be transparent to the user regarding performance. In order to do so, comparisons of CPU, memory, disk I / O, network and application were made, evidencing a low overhead of lightweight virtualization, which comes close to the native scenario in certain cases.

References

Al Jabry, H., Liu, L., Zhu, Y., e Panneerselvam, J. (2014). A critical evaluation of the In Communications and Networking in performance of virtualization technologies. China (CHINACOM), 2014 9th International Conference on, Páginas 606–611. IEEE.

Coker, R. (2001). Bonnie++. http://www.coker.com.au/bonnie++/. Acessado em 15/02/2017.

Docker, I. (2013). Docker Build, Ship, and Run Any App, Anywhere. https://www.docker.com. Acessado em 01/02/2017.

Eiras, R. S., Couto, R. S., e Rubinstein, M. G. (2016). Performance evaluation of a virtualized HTTP proxy in KVM and Docker. In Network of the Future (NOF), 2016 7th International Conference on the, Páginas 1–5. IEEE.

Ismail, B. I., Goortani, E. M., Ab Karim, M. B., Tat, W. M., Setapa, S., Luke, J. Y., e Hoe, In Open Systems O. H. (2015). Evaluation of docker as edge computing platform. (ICOS), 2015 IEEE Confernece on, Páginas 130–135. IEEE.

Kopytov, A. (2004). Sysbench. https://github.com/akopytov/sysbench/. Acessado em 15/02/2017.

Mattos, D. M., Ferraz, L. H. G., Costa, L., e Duarte, O. C. M. (2012a). Virtual network performance evaluation for future internet architectures. Journal of Emerging Technologies in Web Intelligence, 4(4):304–314.

Mattos, D. M., Ferraz, L. H. G., Costa, L. H. M., e Duarte, O. C. (2012b). Evaluating virtual router performance for a pluralist future internet. In Proceedings of the 3rd International Conference on Information and Communication Systems, Página 4. ACM.

McCalpin, J. D. (1991). STREAM: Sustainable Memory Bandwidth in High Performance Computers. https://www.cs.virginia.edu/stream//. Acessado em 15/02/2017.

McMahon, R. (2003). Iperf2. https://sourceforge.net/projects/iperf2/. Acessado em 15/02/2017.

Niu, Z., Xu, H., Tian, Y., Liu, L., Wang, P., e Li, Z. (2016). Benchmarking NFV Software Dataplanes. arXiv preprint arXiv:1605.05843.

Pesic, D., Djordjevic, B., e Timcenko, V. (2016). Competition of virtualized ext4, xfs and btrfs lesystems under type-2 hypervisor. In Telecommunications Forum, 2016 24th, Páginas 1–4. IEEE.

Varghese, B., Subba, L. T., Thai, L., e Barker, A. (2016). Container-Based Cloud Virtual Machine Benchmarking. Páginas 192–201. IEEE.

Yee, A. J. (2009). Y-cruncher A Multi-Threaded Pi-Program. http://www.numberworld.org/y-cruncher/. Acessado em 15/02/2017.