Ashvin Lakshmikantha
Nick McKeown
Abstract
Buffer sizing has received a lot of attention recently since it is becoming increasingly difficult to use large buffers in highspeed routers. Much of the prior work has concentrated on analyzing the amount of buffering required in core routers assuming that TCP carries all the data traffic. In this paper, we evaluate the amount of buffering required for RCP on a single congested link, while explicitly modeling flow arrivals and departures. Our theoretical analysis and simulations indicate that buffer sizes of about 10% of the bandwidth-delay product are sufficient for RCP to deliver good performance to end-users.
- The network simulator: ns-2. Available at http://www.isi.edu/nsnam/ns.Google Scholar
- Rcp web page, 2006. http://yuba.stanford.edu/rcp.Google Scholar
- G. Appenzeller, I. Keslassy, and N. McKeown. Sizing router buffers. In ACM/SIGCOMM, 2004. Google ScholarDigital Library
- H. Balakrishnan, N. Dukkipati, N. McKeown, and C. Tomlin. Stability analysis of explicit congestion control protocols. IEEE Communication Letters, 11(10), October 2007.Google ScholarCross Ref
- T. Bonald, L. Massoulie, A. Proutiere, and J. Virtamo. A queueing analysis of max-min fairness, proportional fairness and balanced fairness. Queueing Systems, 2006. Google ScholarDigital Library
- M. Handley D. Katabi and C. Rohrs. Internet congestion control for future high bandwidth-delay product environments. In Proceedings of ACM SIGCOMM, 2002. Google ScholarDigital Library
- G. de Veciana, G. Kesidis, and J. Walrand. Resource management in wide-area ATM networks using effective bandwidths. IEEE Journal on Selected Areas in Communications, 13:1081--1090, 1995. Google ScholarDigital Library
- A. Dhamdhere and C. Dovrolis. Open issues in router buffer sizing. ACM/SIGCOMM Computer Communication Review, pages 87--92, January 2006. Google ScholarDigital Library
- N. Dukkipati, G. Gibb, N. McKeown, and J. Zhu. Building a RCP (rate control protocol) test network. In Proceedings of Hot Interconnects, August 2007. Google ScholarDigital Library
- N. Dukkipati, M. Kobayashi, R. Zhang-Shen, and N. McKeown. Processor sharing flows in the Internet. In Thirteenth International Workshop on Quality of Service (IWQoS), 2005. Google ScholarDigital Library
- N. Dukkipati and N. McKeown. Why flow-completion time is the right metric for congestion control. In ACM SIGCOMM Computer Communication Review, January 2006. Google ScholarDigital Library
- N. Dukkipati, N. McKeown, and A. Fraser. RCP-AC: Congestion control to make flows complete quickly in any environment. In Proceedings of the IEEE INFOCOM, April 2006.Google ScholarCross Ref
- S. G. Eick, W. A. Massey, and W. Whitt. The physics of the Mt/G/? queue. Operations Research, 41:731--742, 1993. Google ScholarCross Ref
- M. Enachescu, Y. Ganjali, A. Goel, T. Roughgarden, and N. McKeown. Part III: Routers with very small buffers. ACM/SIGCOMM Computer Communication Review, 35(3):7, July 2005. Google ScholarDigital Library
- C. Fulton, S. Q. Li, and C. S. Lim. An ABR feedback control scheme with tracking. In Proceedings of IEEE INFOCOM, pages 806--815, 1997. Google ScholarDigital Library
- A. J. Ganesh, D. Wischik, and N. O'Connell. Big Queues, volume 1838 of Lecture notes in Mathematics. Springer-Berlin, 2004.Google Scholar
- M. Handley and D. Wischik. Congestion, synchronization and buffer size in backbone routers, 2006. preprint.Google Scholar
- S. Kalyanaraman, R. Jain, S. Fahmy, R. Goyal, and B. Vandalore. The ERICA switch algorithm for ABR traffic management in ATM networks, 1997. http://www.cis.ohio-state.edu/~jain/papers.Google Scholar
- F. Kelly, G. Raina, and T. Voice. Stability and fairness of explicit congestion control with small buffers. SIGCOMM Computer Communication Review, pages 51--62, 2008. Google ScholarDigital Library
- A. Lakshmikantha, N. Dukkipati, R. Srikant, N. McKeown, and C.L. Beck. Performance analysis of RCP, 2006. Technical Report, available at http://www.ifp.uiuc.edu/~lkshmknt/rcp.pdf.Google Scholar
- A. Lakshmikantha, R. Srikant, and C. Beck. Impact of file arrivals and departures on buffer sizing in core routers. In Proceedings of IEEE Infocom, 2008.Google ScholarCross Ref
- L. Massoulie. Structural properties of proportional fairness: stability and insensitivity. The Annals of Applied Probability, 2007.Google ScholarCross Ref
- G. Raina, D. Towsley, and D. Wischik. Part II: Control theory for buffer sizing. ACM/SIGCOMM Computer Communication Review, pages 79--82, July 2005. Google ScholarDigital Library
- C. H. Tai, J. Zhu, and N. Dukkipati. Making large scale deployment of RCP practical for real networks. In IEEE INFOCOM Mini-conference, 2008.Google ScholarCross Ref
- W. Whitt. On the heavy-traffic limit theorems for GI/G/∞ queues. Advances in Applied Probability, 14:171--190, 1982.Google ScholarCross Ref
- D. Wischik and N. McKeown. Part I: Buffer sizes for core routers. ACM/SIGCOMM Computer Communication Review, pages 75--78, July 2005. Google ScholarDigital Library
Index Terms
- Buffer sizing results for RCP congestion control under connection arrivals and departures
Recommendations
Efficient congestion avoidance mechanism
LCN '00: Proceedings of the 25th Annual IEEE Conference on Local Computer NetworksIncreasing uncontrolled best-effort traffic deteriorates the ability of TCP to control congestion and is a source of high drop rates. This paper proposes an efficient congestion avoidance mechanism (ECAM) suitable for uncontrolled unicast and multicast ...
Impact of file arrivals and departures on buffer sizing in core routers
Traditionally, it had been assumed that the efficiency requirements of TCP dictate that the buffer size at the router must be of the order of the bandwidth-delay (C × RTT) product. Recently, this assumption was questioned in a number of papers, and the ...
A receiver centric congestion control mechanism
WiCOM'09: Proceedings of the 5th International Conference on Wireless communications, networking and mobile computingWe propose a new congestion control mechanism to improve network performance. It is based on the end-to-end receiver centric method, and adopts both the rate-based and the window-based functions. The available bandwidth is measured at the receiver and ...
Comments