KIPS Transactions on Computer and Communication Systems (정보처리학회논문지:컴퓨터 및 통신 시스템)

Korea Information Processing Society (한국정보처리학회)
권호 표지
DOI QR코드

DOI QR Code

Performance Analysis and Optimization of OpenDaylight Controller in Distributed Cluster Environment

분산 클러스터 환경에서 오픈데이라이트 컨트롤러 성능 분석 및 최적화

  • 이솔이 (충북대학교 정보통신공학부) ;
  • 김태홍 (충북대학교 정보통신공학부) ;
  • 김태준 (충북대학교 정보통신공학부)
  • Received : 2017.04.05
  • Accepted : 2017.07.06
  • Published : 2017.11.30
Download PDF
⟨ Previous Next ⟩

Abstract

OpenDaylight is an SDN (Software Defined Networking) open source framework, which is popular in network fields recently. This paper analyzes the performance of a controller cluster architecture by focusing on distributed datastore and Raft leader election algorithm. In addition, we propose an enhanced version of Raft algorithm in order to improve the performance of distributed datastore by distributing shard leaders over controller cluster. This paper compares the conventional Raft algorithm with the proposed version of the Raft algorithm. Moreover, we compare the performance of distributed datastore according to shard roles such as leader and follower. Experimental results show that Shard leaders provide better performance than followers and Shard updating requests need to be distributed over multiple controllers. So, by using proposed version of Raft algorithm, controller performance can be improved. The details of the experiment results are cleary described.

본 논문에서는 SDN (Software Defined Networking) 오픈소스 프레임워크인 오픈데이라이트(ODL, OpenDaylight) 컨트롤러 클러스터 환경에서 클러스터의 구조를 분석하며 고가용성(High availability)을 지원하는 컨트롤러 클러스터의 동작 방식을 다룬다. 또한 Raft 알고리즘의 리더 선정(Leader Election) 과정을 분석하고 효율적인 시스템 운용을 위한 Leader Election 과정의 개선 방안을 제안한다. 이와 함께 샤드(Shard) 리더와 샤드 팔로어의 성능차이를 제시하고, 기존과 제안 방식의 컨트롤러 클러스터의 성능을 비교 분석한다. 실험의 결과에 따르면 리더의 성능은 팔로어의 성능보다 좋으며 하나의 컨트롤러로 요청이 집중되어 전달될 때보다 분산된 컨트롤러로 요청이 전달될 때의 성능이 더 좋다. 따라서 제안 기법을 통하여 컨트롤러로의 요청을 분산함으로써 성능을 높일 수 있다.

Keywords

References

  1. J. Y. Lee, "Software Defined Network (SDN)," Retrieved Dec., 2016, from http://www.bloter.net/archives/267815.
  2. D. E. Suh et al., "Optimal Master Controller Assignment for Minimizing Flow Setup Latency in SDN," Computer Communications Workshops (INFOCOM WKSHPS), 2016 IEEE Conference on. IEEE, pp.421-422, 2016.
  3. T. H. Kim et al., "Performance Evaluation and Optimal Operation Strategy of OpenDaylight Controller Cluster," J. KICS, Vol.41, No.12, pp.1801-1810, Dec., 2016. https://doi.org/10.7840/kics.2016.41.12.1801
  4. Colin Dixon, "Clustering in OpenDaylight," OpenDaylight Mini-Summit, Santa Clara, USA, Mar., 2016.
  5. Y. H. Goo et al., "Data Processing Performance Evaluation of ODL Controller in a SDN Controller Cluster Environment," Proceedings of Symposium of the Korean Institute of communications and Information Sciences, pp.1208-1209, Jan., 2016.
  6. D. E. Suh, S. K. Jang, S. Han, S. H. Pack, T. H. Kim, and J. Y. Kwak, "On performance of OpenDaylight clustering," Netsoft 2016, Seoul, Korea, Jun., 2016.
  7. W. S. Jung et al., "Performance Evaluation of ODL High Availability on the Distributed Controller Cluster Environment," Proceedings of Symposium of the Korean Institute of Communications and Information Sciences, pp. 258-259, Nov., 2015.
  8. S. Han et al., "A Study on OpenDaylight Distributed Controller Architecture," Proceedings of Symposium of the Korean Institute of communications and Information Sciences, pp. 337-338, Jan., 2016.
  9. Dixit, Advait et al., "Towards an elastic distributed SDN controller," ACM SIGCOMM Computer Communication Review, ACM, Vol.43. No.4. pp.7-12, 2013.
  10. J. W. Kyung et al., "A load distribution scheme over multiple controllers for scalable SDN," Ubiquitous and Future Networks (ICUFN), 2015 Seventh International Conference on. IEEE, pp.808-810, 2015.
  11. S. E. Lee et al., "A CPU Load-Based Master Controller Election Scheme for Distributed SDN Controllers," Proceedings of Symposium of the Korean Institute of communications and Information Sciences, pp.215-216, Jan., 2017.
  12. Diego Ongaro et al., "In Search of an Understandable Consensus Algorithm," USENIX Annual Technical Conference 2014, Philadelphia, USA, Jun., 2014.
  13. The Opendaylight Project/controller, Retrieved Dec., 2016, from https://github.com/opendaylight/controller.
  14. The Raft Consensus Algorithm, Retrieved Dec., 2016, from https://raft.github.io.
  15. Ongaro, Diego, "Consensus: Bridging theory and practice," Diss. Stanford University, 2014.
  16. OpenDaylight Project, Retrieved Dec., 2016, from https://www.oepndaylight.org.
  17. K. B. Noh et al., "A Study of Software Defined Networking Migration Method," Entrue Journal of Information Technology, Vol.13, No.3, pp.35-58, Dec., 2014.
  18. J. H. You, W. S. Kim, and C. H. Yoon, "A Technical Trend and Prospect of Software Defined Network and OpenFlow," KNOM Review, Vol.15, No.2, pp.1-24, Dec., 2012.
  19. Akka, Retrieved Dec., 2016, from https://www.akka.io.
  20. Medved, Jan et al., "Opendaylight: Towards a model-driven sdn controller architecture," A World of Wireless, Mobile and Multimedia Networks (WoWMoM), 2014 IEEE 15th International Symposium on. IEEE, pp.1-6, 2014.
  21. Moiz Raja, "MD-SAL Clustering Internals," OpenDaylight Summit 2015, Santa Clara, USA, Jul., 2015.
  22. Neelakrishnan, Priyanka., "Enhancing scalability and performance in software-defined networks: An OpenDaylight (ODL) case study," Diss. San Jose State University, 2016.