The Transactions of the Korea Information Processing Society (정보처리학회 논문지)

Korea Information Processing Society (한국정보처리학회)
권호 표지
DOI QR코드

DOI QR Code

Performance Evaluation and Analysis on Single and Multi-Network Virtualization Systems with Virtio and SR-IOV

가상화 시스템에서 Virtio와 SR-IOV 적용에 대한 단일 및 다중 네트워크 성능 평가 및 분석

  • 이재학 (고려대학교 정보창의교육연구소) ;
  • 임종범 (평택대학교 스마트콘텐츠학과 ) ;
  • 유헌창 (고려대학교 컴퓨터학과 )
  • Received : 2023.10.19
  • Accepted : 2024.01.20
  • Published : 2024.02.29
Download PDF
⟨ Previous Next ⟩

Abstract

As functions that support virtualization on their own in hardware are developed, user applications having various workloads are operating efficiently in the virtualization system. SR-IOV is a virtualization support function that takes direct access to PCI devices, thus giving a high I/O performance by minimizing the need for hypervisor or operating system interventions. With SR-IOV, network I/O acceleration can be realized in virtualization systems that have relatively long I/O paths compared to bare-metal systems and frequent context switches between the user area and kernel area. To take performance advantages of SR-IOV, network resource management policies that can derive optimal network performance when SR-IOV is applied to an instance such as a virtual machine(VM) or container are being actively studied.This paper evaluates and analyzes the network performance of SR-IOV implementing I/O acceleration is compared with Virtio in terms of 1) network delay, 2) network throughput, 3) network fairness, 4) performance interference, and 5) multi-network. The contributions of this paper are as follows. First, the network I/O process of Virtio and SR-IOV was clearly explained in the virtualization system, and second, the evaluation results of the network performance of Virtio and SR-IOV were analyzed based on various performance metrics. Third, the system overhead and the possibility of optimization for the SR-IOV network in a virtualization system with high VM density were experimentally confirmed. The experimental results and analysis of the paper are expected to be referenced in the network resource management policy for virtualization systems that operate network-intensive services such as smart factories, connected cars, deep learning inference models, and crowdsourcing.

하드웨어 자체적으로 가상화를 지원하는 기능들이 추가됨에 따라 다양한 작업 유형을 가진 사용자 어플리케이션들이 가상화 시스템에서 효율적으로 운용되고 있다. 가상화 지원 기능 중 SR-IOV는 PCI 장치에 대한 직접 접근을 통해 하이퍼바이저 또는 운영체제 개입을 최소화하여 시스템 성능을 높이는 기술로 베어-메탈 시스템 대비 비교적 긴 I/O 경로 및 사용자 영역과 커널 영역에 대한 빈번한 컨텍스트 스위칭 등 가상화 계층의 추가로 낮은 네트워크 성능을 가진 가상화 시스템에서 네트워크 I/O 가속화를 실현하게 해준다. 이러한 성능적 이점을 이용하기 위해 가상머신 또는 컨테이너와 같은 인스턴스에 SR-IOV를 접목할 시 최적의 네트워크 I/O 성능을 도출할 수 있는 네트워크 자원 관리 정책이 활발히 연구되고 있다. 본 논문은 I/O 가속화를 실현하는 SR-IOV의 네트워크 성능을 1) 네트워크 지연 시간, 2) 네트워크 처리량, 3) 네트워크 공정성, 4) 성능간섭, 5) 다중 네트워크와 같은 측면으로 세밀한 성능 평가 및 분석을 Virtio와 비교하여 진행한다. 본 논문의 기여점은 다음과 같다. 첫째, 가상화 시스템에서 Virtio와 SR-IOV의 네트워크 I/O 과정을 명확히 설명했으며, 둘째, Virtio와 SR-IOV의 네트워크 성능을 다양한 성능 메트릭을 기반으로 분석하였다. 셋째, 가상머신 밀집도가 높은 환경에서 SR-IOV 네트워크에 대한 시스템 오버헤드 및 이에 대한 최적화 가능성을 실험으로 확인하였다. 본 논문의 실험 결과 및 분석들은 스마트 팩토리, 커넥티드-카, 딥러닝 추론 모델, 크라우드 소싱과 같은 네트워크 집약적인 서비스들을 운용하는 가상화 시스템에 대한 네트워크 자원 관리 정책에 활용될 것으로 기대된다.

Keywords

Acknowledgement

본 연구는 산림청(한국임업진흥원) 산림과학기술 연구개발사업 '2022427C10-2224-0801'의 지원에 의하여 이루어진 것임.

References

  1. Y. Dong, X. Yang, J. Li, G. Liao, K. Tian, and H. Guan, "High performance network virtualization with SR-IOV," Journal of Parallel and Distributed Computing, Vol.72, No.11, pp.1471-1480, 2012.  https://doi.org/10.1016/j.jpdc.2012.01.020
  2. G. Motika and S. Weiss, "Virtio network paravirtualization driver: Implementation and performance of a de-facto standard," Computer Standards & Interfaces, Vol.34, No.1, pp.36-47, 2012.  https://doi.org/10.1016/j.csi.2011.05.002
  3. X. Hu, J. Li, R. Ma, and H. Guan, "ES2: Building an efficient and responsive event path for I/O virtualization," IEEE Transactions on Cloud Computing, Vol.10, No.2, pp.1358-1372, 2020.  https://doi.org/10.1109/TCC.2020.2969660
  4. J. Nakajima et al., "Optimizing virtual machines using hybrid virtualization," In: Proceedings of the 2011 ACM Symposium on Applied Computing, pp.573-578, 2011. 
  5. Y. Dong, X. Zhang, J. Dai, and H. Guan, "HYVI: a hybrid virtualization solution balancing performance and manageability," IEEE Transactions on Parallel and Distributed Systems, Vol.25, No.9, pp.2332-2341, 2013.  https://doi.org/10.1109/TPDS.2013.239
  6. S. C. Mhatre and P. Chandran, "On making xen detect hypercalls and memory accesses for simulating virtualization-enabled processors," In: Proceedings of the 35th Annual ACM Symposium on Applied Computing, pp.154-161, 2020. 
  7. Y. Dong, Z. Yu, and G. Rose, "SR-IOV Networking in Xen: Architecture, Design and Implementation," In: Workshop on I/O virtualization, 2008. 
  8. I. Cerrato, M. Annarumma, and F. Risso, "Supporting fine-grained network functions through Intel DPDK," In: 2014 Third European Workshop on Software Defined Networks, IEEE. pp.1-6, 2014. 
  9. H. Fayyad-Kazan, L. Perneel, and M. Timmerman, "Full and para-virtualization with Xen: a performance comparison," Journal of Emerging Trends in Computing and Information Sciences, Vol.4, No.9, pp.719-727, 2013. 
  10. S. Sultan, I. Ahmad, and T. Dimitriou, "Container security: Issues, challenges, and the road ahead," IEEE Access, Vol.7, pp.52976-52996, 2019.  https://doi.org/10.1109/ACCESS.2019.2911732
  11. M. Belair, S. Laniepce, and J.-M. Menaud, "Leveraging kernel security mechanisms to improve container security: a survey," In: Proceedings of the 14th International Conference on Availability, Reliability and Security, pp.1-6, 2019. 
  12. K. Brady, S. Moon, T. Nguyen, and J. Coffman, "Docker container security in cloud computing," In: 2020 10th Annual Computing and Communication Workshop and Conference (CCWC), IEEE. pp.0975-0980, 2020. 
  13. A. Kivity, Y. Kamay, D. Laor, U. Lublin, and A. Liguori, "kvm: the Linux virtual machine monitor," In: Proceedings of the Linux symposium, pp.225-230, 2007. 
  14. F. Bellard, "QEMU, a fast and portable dynamic translator," In: USENIX Annual Technical Conference, FREENIX Track. pp.46, 2005. 
  15. ping, Accessed in September, 2023. [Internet], https://manpages.ubuntu.com/manpages/jammy/en/man1/ping.1.html 
  16. stress, Accessed in September, 2023. [Internet], http://manpages.ubuntu.com/manpages/jammy/en/man1/stress.1.htmls 
  17. iperf, Accessed in September, 2023. [Internet], http://manpages.ubuntu.com/manpages/jammy/en/man1/iperf.1.html 
  18. W. Jia, J. Shan, TO. Li, X. Shang, H. Cui, and X. Ding, "{vSMT-IO}: Improving {I/O} Performance and Efficiency on {SMT} Processors in Virtualized Clouds," In: 2020 USENIX Annual Technical Conference (USENIX ATC 20), pp.449-463, 2020. 
  19. K. Suo, Y. Zhao, J. Rao, L. Cheng, X. Zhou, and FCM. Lau, "Preserving i/o prioritization in virtualized oses," In: Proceedings of the 2017 Symposium on Cloud Computing, p.269-281, 2017. 
  20. S. G. Soriga and B. Mihai, "A comparison of the performance and scalability of Xen and KVM hypervisors," 2013 RoEduNet International Conference 12th Edition: Networking in Education and Research, IEEE, 2013. 
  21. G. Motika and W. Shlomo, "Virtio network paravirtualization driver: Implementation and performance of a de-facto standard," Computer Standards & Interfaces, Vol.34, No.1, pp.36-47, 2012.  https://doi.org/10.1016/j.csi.2011.05.002
  22. S. A. Babu, M. J. Hareesh, J. P. Martin, S. Cherian, and Y. Sastri, "System performance evaluation of para virtualization, container virtualization, and full virtualization using xen, openvz, and xenserver," In 2014 Fourth International Conference on Advances in Computing and Communications, IEEE, pp.247-250, 2014. 
  23. M. Fischer and W. Florian, "Survey on SR-IOV performance," Network, Vol.43, 2021. 
  24. A. T. de Oliveira Filho, E. Freitas, PRX. do. Carmo, DFH. Sadok, and J. Kelner, "Measuring the impact of SR-IOV and virtualization on packet round-trip time," Computer Communications, Vol.211, pp.193-215, 2023.  https://doi.org/10.1016/j.comcom.2023.09.013
  25. R. Shea and L. Jiangchuan, "Network interface virtualization: challenges and solutions," IEEE Network, Vol.26, No.5, pp.28-34, 2012.  https://doi.org/10.1109/MNET.2012.6308072
  26. J. Liu, "Evaluating standard-based self-virtualizing devices: A performance study on 10 GbE NICs with SR-IOV support," 2010 IEEE International Symposium on Parallel & Distributed Processing (IPDPS), IEEE, 2010. 
  27. R. Ganesan, Y. Murarka, S. Sarkar, and K. Frey, "Empirical study of performance benefits of hardware assisted virtualization," Proceedings of the 6th ACM India Computing Convention, 2013.