Deficit Round Robin with Limited Deficit Savings (DRR-LDS) for Fairness Among TCP Users

Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 10740)

Abstract

Deficit Round Robin (DRR) is a simple and computationally efficient approximation of the Weighted Fair Queueing (WFQ) scheduling discipline. Its intention is to share resources among several queues, e.g., flows or users, according to given weights. However, when users hold different numbers of TCP connections with saturated sources, the throughput among these users may differ significantly.

In this work, we quantify the difference in throughput for heavy and light users with saturated TCP flows for equal weights and for two different buffer management strategies. The difference is large if low queueing delay for packets is enforced through shallow buffers on the bottleneck link. To address this shortcoming, we propose limited deficit savings (LDS), a modification of the DRR scheduler, which can be combined with different buffer management schemes. We show that LDS reduces unequal throughput for heavy and light users with saturated TCP flows. Moreover, we illustrate that LDS clearly decreases download times for data chunks of moderate size in the presence of high background load.

Keywords

Congestion management Scheduling Fairness Buffer management 

References

  1. 1.
    FP7 Project (STREP): Reducing Internet Transport Latency (RITE), November 2012–2015Google Scholar
  2. 2.
    Adams, R.: Active queue management: a survey. IEEE Commun. Surv. Tutor. 15(3), 1425–1476 (2013)CrossRefGoogle Scholar
  3. 3.
    Allman, M.: Comments on bufferbloat. ACM SIGCOMM Comput. Commun. Rev. 43(1), 30–37 (2013)CrossRefGoogle Scholar
  4. 4.
    Bennett, J.C., Zhang, H.: WF\(^2\)Q : worst-case fair weighted fair queueing. In: IEEE Infocom (1996)Google Scholar
  5. 5.
    Briscoe, B., Brunstrom, A., Petlund, A., Hayes, D., Ros, D., Tsang, I.J., Gjessing, S., Fairhurst, G., Griwodz, C., Welzl, M.: Reducing internet latency: a survey of techniques and their merits. IEEE Commun. Surv. Tutor. 18(3), 2149–2196 (2016)CrossRefGoogle Scholar
  6. 6.
    Demers, A., Keshav, S., Shenker, S.: Analysis and simulation of a fair queuing algorithm. In: ACM SIGCOMM (1989)Google Scholar
  7. 7.
    Dhamdhere, A., Dovrolis, C.: Open issues in router buffer sizing. ACM SIGCOMM Comput. Commun. Rev. 36(1), 87–92 (2006)CrossRefGoogle Scholar
  8. 8.
    Dhamdhere, A., Jiang, H., Dovrolis, C.: Buffer sizing for congested internet links. In: IEEE Infocom. Miami, FL, March 2005Google Scholar
  9. 9.
    Floyd, S., Jacobson, V.: Random early detection gateways for congestion avoidance. IEEE/ACM Trans. Netw. 1(4), 397–413 (1993)CrossRefGoogle Scholar
  10. 10.
    Gettys, J., Nichols, K.: Bufferbloat: dark buffers in the internet. ACM Queue 9(11), 40 (2011)CrossRefGoogle Scholar
  11. 11.
    Gong, Y., Rossi, D., Leonardi, E.: Modeling the interdependency of low-priority congestion control and active queue management. In: International Teletraffic Congress (ITC) (2013)Google Scholar
  12. 12.
    Gong, Y., Rossi, D., Testa, C., Valenti, S., Taht, M.D.: Fighting the bufferbloat: on the coexistence of AQM and low priority congestion control. In: IEEE INFOCOM Workshop on Traffic Measurement and Analysis (2013)Google Scholar
  13. 13.
    Hoeiland-Joergensen, T., McKenney, P., Taht, D., Gettys, J., Dumazet, E.: RFC8290: The Flow Queue CoDel Packet Scheduler andActive Queue Management Algorithm, January 2018. https://tools.ietf.org/html/rfc8290
  14. 14.
    Hohlfeld, O., Pujol, E., Ciucu, F., Feldmann, A., Barford, P.: BufferBloat: How Relevant? A QoE Perspective on Buffer Sizing. Technical report 2012–11, TU Berlin, Faculty of Electrical Engineering and Computer Science, November 2012Google Scholar
  15. 15.
    IETF Working Group on Active Queue Management and Packet Scheduling (AQM): Description of the Working Group (2013). http://tools.ietf.org/wg/aqm/charters
  16. 16.
    IETF Working Group on Congestion Exposure (CONEX): Description of the Working Group (2010). http://tools.ietf.org/wg/conex/charters
  17. 17.
    IETF Working Group on RTP Media Congestion Avoidance Techniques (RMCAT): Description of the Working Group (2012). http://tools.ietf.org/wg/rmcat/charters
  18. 18.
    Internet Society: Bandwidth Management - Internet Society Technology Roundtable Series, November 2012. http://www.internetsociety.org/sites/default/files/BWroundtable_report-1.0.pdf
  19. 19.
    Internet Society: Report on the Workshop on Reducing Internet Latency, December 2013. http://www.internetsociety.org/latency2013
  20. 20.
    Jiang, H., Liu, Z., Wang, Y., Lee, K., Rhee, I.: Understanding bufferbloat in cellular networks. In: Workshop on Cellular Networks: Operations, Challenges, and Future Design (CellNet), August 2012Google Scholar
  21. 21.
    Khademi, N., Ros, D., Welzl, M.: The New AQM Kids on the Block: Much Ado About Nothing? Technical report 434, University of Oslo, Department of Informatics (2013)Google Scholar
  22. 22.
    McKenney, P.E.: Stochastic fairness queueing. In: IEEE Infocom (1990)Google Scholar
  23. 23.
    Nagle, J.: RFC970: On Packet Switches with Infinite Storage, December 1985Google Scholar
  24. 24.
    Nichols, K., Jacobson, V., McGregor, Ed., A., Iyengar, Ed., J.: RFC8289: Controlled Delay Active Queue Management, January 2018. https://tools.ietf.org/html/rfc8289
  25. 25.
    Nichols, K., Jacobson, V.: Controlling queue delay. ACM Queue 10(5), 1–15 (2012)CrossRefGoogle Scholar
  26. 26.
    Pan, R., Natarajan, P., Piglione, C., Prabhu, M.S., Subramanian, V., Baker, F., VerSteeg, B.: PIE: a lightweight control scheme to address the bufferbloat problem. In: IEEE Workshop on High Performance Switching and Routing (HPSR) (2013)Google Scholar
  27. 27.
    Ramakrishnan, K., Floyd, S., Black, D.: RFC3168: The Addition of Explicit Congestion Notification (ECN) to IP, September 2001Google Scholar
  28. 28.
    Shreedhar, M., Varghese, G.: Efficient fair queuing using deficit round robin. IEEE/ACM Trans. Netw. 4(3), 375–385 (1996)CrossRefGoogle Scholar
  29. 29.
    Suter, B., Lakshman, T.V., Stiliadis, D., Choudhury, A.K.: Buffer management schemes for supporting TCP in gigabit routers with per-flow queueing. IEEE J. Sel. Areas Commun. 17(6), 1159–1169 (1999)CrossRefGoogle Scholar
  30. 30.
    Varga, A.: INET-2.2 released, August 2013. http://inet.omnetpp.org/
  31. 31.
    Varga, A.: OMNeT++ 4.3.1 released, September 2013. http://www.omnetpp.org/
  32. 32.
    Wand, S.: Network Simulation Cradle (2012). http://research.wand.net.nz/software/nsc.php
  33. 33.
    White, G., Pan, R.: RFC8034: Active Queue Management (AQM) Based on Proportional Integral Controller Enhanced (PIE) for Data-Over-Cable Service Interface Specifications (DOCSIS) Cable Modems, February 2017. https://tools.ietf.org/html/rfc8034
  34. 34.
    White, G.: Data-over-Cable Service Interface Specification - MAC and Upper Layer Protocols Interface Specification. Technical report CM-SP-MULPIv3.1-I01-131029, Cable Television Laboratories, Inc., October 2013Google Scholar
  35. 35.
    White, G.: Active Queue Management in Docsis 3.X Cable Modems. Technical report, Cable Television Laboratories, Inc., May 2014Google Scholar
  36. 36.
    Zhang, L.: VirtualClock: a new traffic control algorithm for packet switching networks. In: ACM SIGCOMM (1990)Google Scholar

Copyright information

© Springer International Publishing AG 2018

Authors and Affiliations

  1. 1.Chair of Communication NetworksUniversity of TuebingenTuebingenGermany

Personalised recommendations