An Analytical Traffic Model of Control Plane and Application Plane in Software-Defined Networking based on Queuing Theory

대기행렬 이론 기반 SDN 제어 평면 및 응용 평면의 트래픽 성능 분석 모델

  • 이승운 (아주대학교 컴퓨터공학과) ;
  • 노병희 (아주대학교 컴퓨터공학과)
  • Received : 2019.08.01
  • Accepted : 2019.08.20
  • Published : 2019.08.31

Abstract

Software Defined Networking (SDN) is the future network paradigm of decoupling control and data functions. In SDN structure, it is hard to address scalability in case of large-scale networks because single controller managed thousands of switches in a centralized fashion. Most of previous studies have focused on horizontal scalability, where distributed controllers are assigned to network devices. However, they have abstracted the control plane and the application plane into a single controller. The layer of the common SDN architecture is divided into data plane, control plane, and application plane, but the control plane and application plane have been modeled as a single controller although they are logically separated. In this paper, we propose a analytical traffic model considering the both application plane and control plane based on queuing theory. This model can be used to address scalability issues such as controller placement problem without complicated simulations.

소프트웨어 정의 네트워킹(SDN)은 네트워크 장치의 데이터 기능과 제어기능을 분리한 미래 네트워킹 기술이다. 네트워크 장치들이 하나의 컨트롤러에 의해 중앙 제어되는 SDN의 구조는 대규모의 네트워크로의 확장성을 보장하기가 쉽지 않다. 선행연구들은 다수의 컨트롤러를 배치하는 수평적 확장성에 초점을 두어왔다. 이 연구들은 제어 평면과 응용 평면을 하나의 컨트롤러로 추상화하고 있다. 보편적인 SDN 아키텍처의 계층은 데이터 평면, 제어 평면, 응용 평면으로 나뉘는데 응용평면과 제어평면은 논리적으로 분리되어 있음에도 두 평면을 하나의 컨트롤러로 간주하였다. 따라서 본 논문에서는 대기행령 이론을 바탕으로 제어 평면과 응용 평면을 분리하여 반영한 트래픽 성능 분석 모델을 제안한다. 이 모델을 사용하면 복잡한 시뮬레이션 대신 간단하게 컨트롤러 배치 문제 등과 같은 확장성 문제에 활용할 수 있다.

Keywords

References

  1. Software-Defined Networking: The New Norm for Networks, Technical Report. Open Networking Foundation (ONF), 2012.
  2. A. Tootoonchian and Y. Ganjali, "HyperFlow: A Distributed Control Plane for Openflow,"I NM/WREN'10, Apr. 2010.
  3. A. Dixit, F. Hao, S. Mukherjee, T. V. Lakshman, and R. Kompella, "Towards an elastic distributed SDN controller," HotSDN'13, pp. 7-12, Hong Kong, China, Aug. 2013.
  4. S. Yeganeh, Y. Ganjali, "Kandoo: A framework for efficient and scalable offloading of control applications", Proc. 1st Workshop HotSDN, pp. 19-24, 2012.
  5. OpenDaylight project, Retrieved Aug. 26, 2016, from www.opendaylight.org
  6. P. Berde, et al., "ONOS: towards an open, distributed SDN OS," HotSDN'14, pp. 1-6, Chicago, Illinois, Aug. 2014.
  7. A. Tavakoli, et al. "Applying NOX to the Datacenter." Proc. HotNets, Oct. 2009.
  8. T. Benson, A. Akella, and D. Maltz, "Network traffic characteristics of data centers in the wild," In Proc. IMC, 2010
  9. ONOS whitepaper, "Raising the bar on SDN control plane performance and scalability," ON.Lab, April, 2015
  10. S. Rehman, W. Song, M. Kang, "Network-wide traffic visibility in OF@TEIN SDN testbed using sFlow," IEEE APNOMS, pp. 1-6, Sep. 2014.
  11. J. Suarez-Varela, P. Barlet-Ros, "Towards a NetFlow implementation for OpenFlow Software-Defined Networks." In 29th IEEE ITC, vol. 1, pp. 187-195, 2017.
  12. R. Jang, D. Cho, Y. Noh and D. Nyang, "RFlow+: An SDN-based WLAN monitoring and management framework," IEEE INFOCOM 2017 - IEEE Conference on Computer Communications, Atlanta, GA, 2017, pp. 1-9. 2017.
  13. M. Hayes, B. Ng, A. Pekar and W. K. G. Seah, "Scalable Architecture for SDN Traffic Classification," in IEEE Systems Journal, vol. 12, no. 4, pp. 3203-3214, Dec. 2018. https://doi.org/10.1109/jsyst.2017.2690259
  14. 황제승, 김웅수, 박준석, 염근혁, "SDN 환경에서 온톨로지를 활용한 애플리케이션 네트워크의 품질 위반 상황 식별 방법," 한국차세대컴퓨팅학회 논문지, 13권 6호, 2017.
  15. 이승운,노병희, "SDN 모델링 및 시뮬레이션을 위한 Mininet과 OPNET 간 성능 비교 분석," 한국차세대컴퓨팅학회 논문지, 14권 2호, 82-93, 2018.
  16. T. Hu, Z. Guo, P. Yi, T. Baker, J. Lan, "Multi-controller Based Software-Defined Networking: A Survey," in IEEE Access, vol. 6, pp. 15980-15996, 2018. https://doi.org/10.1109/ACCESS.2018.2814738
  17. G. Wang, Y. Zhao, J. Huang, Y. Wu, "An Effective Approach to Controller Placement in Software Defined Wide Area Networks," in IEEE Transactions on Network and Service Management, vol. 15, no. 1, pp. 344-355, March 2018. https://doi.org/10.1109/tnsm.2017.2785660