A Study of Core-Stateless Mechanism for Fair Bandwidth Allocation

대역 공평성 보장을 위한 Core-Stateless 기법 연구

  • 김화숙 (한국전자통신연구원 네트워크연구소 네트워크구조팀) ;
  • 김상화 (충남대학교 컴퓨터과학과 내트워크 연구실) ;
  • 김영부 (한국전자통신연구원 네트워크연구소 네트워크구조팀)
  • Published : 2003.04.01


Fair bandwidth allocations at routers protect adaptive flows from non-adaptive ones and may simplify end-to end congestion control. However, traditional fair bandwidth allocation mechanisms, like Weighted Fair Queueing and Flow Random Early Drop, maintain state, manage buffera and perform packet scheduling on a per-flow basis. These mechanisms are more complex and less scalable than simple FIFO queueing when they are used in the interi or of a high-speed network. Recently, to overcome the implementation complexity problem and address the scalability and robustness, several fair bandwidth allocation mechanisms without per-flow state in the interior routers are proposed. Core-Stateless Fair Queueing and Rainbow Fair Queuing are approximates fair queueing in the core-stateless networks. In this paper, we proposed simple Layered Fair Queueing (SLFQ), another core-stateless mechanism to approximate fair bandwidth allocation without per-flow state. SLFQ use simple layered scheme for packet labeling and has simpler packet dropping algorithm than other core-stateless fair bandwidth allocation mechanisms. We presente simulations and evaluated the performance of SLFQ in comparison to other schemes. We also discussed other are as to which SLFQ is applicable.

라우터에서 공평 대역 할당은 네트워크 폭주 상황에 반응하여 흐름을 제어하는 트래픽을 그렇지 않은 트래픽으로부터 보호한다. 그러나 Weighted Fair Queueing, Flow Random Early Drop 등의 전통적인 공평 대역 할당 방법은 상태 관리나 버퍼 관리, 스케쥴링 등을 흐름 단위로 수행하므로 초고속 네트워크의 백본에서 사용되는 경우 구현의 복잡성 뿐 아니라 흐름의 증가에 따른 네트위크의 확장성과 견고성을 확보하기 위해 코어 라우터에서의 흐름단위 관리를 배제하는 Core-stateless 메커니즘들이 제안되었다. Core-Stateless Fair Queueing 과 Rainbow Fair Queueing 등은 Core-Stateless 네트워크에서 근접한 공평 대역 할당 메커니즘들에 비해 단순해진 패킷 레이블 할당 방법과 패킷 폐기 방법을 사용하여 실제 구현 가능성을 높인 새로운 Core-stateless 알고리즘인 Simple Layered Fair Queueing (SLFQ) 를 제안하고 몇가지 형태의 시뮬레이션과 그 결과를 다른 메커니즘의 결과와 비교하면서 대역 공평성 보장 정도를 확인한다. 마지막으로 제안된 알고리즘의 향후 응용 가능성을 제시한다.



  1. S.Floyd, 'Promoting the Use of End-to-End Congestion in the Internet,' IEEE/ACM Tr. on Networking, 7(4), pp.458-472, Aug. 1999 https://doi.org/10.1109/90.793002
  2. A-Demers, et al., 'Analysis and Simulation of a Fair Queueing Algorithm,' Proceedings of ACM SIGCOMM'89, pp.3-12, Sep. 1989
  3. Done Lin, et al., Dynamics of Random Early Detection,' In Proceedings of ACM SIGCOMM'97, 00.127-137, Oct. 1997
  4. Ion Stoica, et al., ' Core-Stateless Fair Que ueing: Achieving Approximately Fair Bandw idth Allocations in High Speed Networks,'Proceedings of ACM SIGCOMM'98, pp.118-130, 1998
  5. Ziruo Cao, et al., 'Rainbow Fair Queueing: F air Bandwidth Sharing without Per-Flow St ate,' Proceedings cf IEEE INFOCOM 2000, Vol. 2, pp. 922 -931, 2000 https://doi.org/10.1109/INFCOM.2000.832267
  6. Zhai Mingyu, et al., ' Fair bandwidth allocat ions through queue management in core-stat eless networks,' Proceedings cf IEEE GLO BECOM2001. Vol. 4. pp.2248-2252 2001
  7. Takashi Kunmoto, et al., ' Core-Stateless RED Algoiithm for Improving Faimess in a Best-Effort Network,' IEICE Tr. Com, E84-B(5), PP. 1413-1422, May 2001
  8. lon Stoica, et al., 'Providing Guaranteed Ser vices Without Per Flow Management', Proc eedings cf SIGCOMM'99, pp.81-94, September 1999
  9. V. Eleck, et al., 'Admission Control Based on End-to-End Measurements, Proceedings of IEEE INFOCOM2000, pp.623-630, 2000 https://doi.org/10.1109/INFCOM.2000.832236
  10. Woo-Seop Rhee, et al., 'Two phase edge-to -edge distributed measurement based admissi on control mechanism in large IP networks,'GLOBECOM'01, Vo1.4, pp.2571-2575, 2001
  11. Antoine Clerget, et al.,'TUF: Tag-based Unified Faimess,' Proceedings cf IEEE IN FOCOM2001, Vo1.1, pp.498-507, 2001
  12. Cellio Albuquerque, et al., 'Network Border Patrol,' Proceedings cf IEEE INFOCOM2000, Vol.l, pp.322-331, 2000
  13. M.Shreedhar, et al., 'Efficient Fair Queueing using Deficit Round Robin,' Proceedings of SIGC0MM'95, pp.231-243, Sep. 1995
  14. Sally Floyd, et al., 'Random Early Detection for Congestion Avoidance,' IEEE/ACM Transaction on Networking, 1(4), PP.397-413,Tuly 1993 https://doi.org/10.1109/90.251892
  15. UCB/LBNLMNT, 'Network Simulator - NS (Version 2),' http://www.isi.edu/nsnam/ns/
  16. Ion Stoica, 'NS-2 Simulation for CSFQ,' http://www.cs.cmu.edu/~istoica/csfq, 1998