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 (C)-delay (RTT) product. Recently this assumption was questioned in a number of papers and the rule was shown to be conservative for certain traffic models. In particular, by appealing to statistical multiplexing it was shown that on a router with N long-lived connections, buffers of size O(CxRTT)/radic(N) or even O(1) are sufficient. In this paper, we reexamine the buffer size requirements of core routers when flows arrive and depart. Our conclusion is as follows: if the core to access speed ratio is large, then O(1) buffers are sufficient at the core routers; otherwise, larger buffer sizes do improve the flow-level performance of the users. From a modeling point of view, our analysis offers two new insights. First, it may not be appropriate to derive buffer-sizing rules by studying a network with a fixed number of users. In fact, depending upon the core-to-access speed ratio, the buffer size itself may affect the number of flows in the system, so these two parameters (buffer size and number of flows in the system) should not be treated as independent quantities. Second, in the regime where the core-to- access speed ratio is large, we note that the O(1) buffer sizes are sufficient for good performance and that no loss of utilization results, as previously believed.

[1]  Frank Kelly,et al.  Reversibility and Stochastic Networks , 1979 .

[2]  Nick McKeown,et al.  Part I: buffer sizes for core routers , 2005, CCRV.

[3]  Donald F. Towsley,et al.  Part II: control theory for buffer sizing , 2005, CCRV.

[4]  Jin Cao,et al.  A Poisson limit for buffer overflow probabilities , 2002, Proceedings.Twenty-First Annual Joint Conference of the IEEE Computer and Communications Societies.

[5]  Cheng Song,et al.  High performance TCP in ANSNET , 1994, CCRV.

[6]  Fernando Paganini,et al.  FAST TCP: from theory to experiments , 2005, IEEE Network.

[7]  Murad S. Taqqu,et al.  A Practical Guide to Heavy Tails: Statistical Techniques for Analysing Heavy-Tailed Distributions , 1998 .

[8]  Ward Whitt,et al.  The Brownian approximation for rate-control throttles and the G/G/1/C queue , 1992, Discret. Event Dyn. Syst..

[9]  Donald F. Towsley,et al.  Modeling TCP throughput: a simple model and its empirical validation , 1998, SIGCOMM '98.

[10]  R. Srikant,et al.  Diffusion approximations for a single node accessed by congestion-controlled sources , 2000, IEEE Trans. Autom. Control..

[11]  Tim Roughgarden,et al.  Part III: routers with very small buffers , 2005, CCRV.

[12]  Thomas Bonald,et al.  Statistical bandwidth sharing: a study of congestion at flow level , 2001, SIGCOMM.

[13]  Amogh Dhamdhere,et al.  Open issues in router buffer sizing , 2006, CCRV.

[14]  Chase Cotton,et al.  Packet-level traffic measurements from the Sprint IP backbone , 2003, IEEE Netw..

[15]  Hao Jiang,et al.  Buffer sizing for congested Internet links , 2005, Proceedings IEEE 24th Annual Joint Conference of the IEEE Computer and Communications Societies..

[16]  Azer Bestavros,et al.  Self-similarity in World Wide Web traffic: evidence and possible causes , 1997, TNET.

[17]  Dimitri P. Bertsekas,et al.  Data Networks , 1986 .

[18]  Injong Rhee,et al.  Binary increase congestion control (BIC) for fast long-distance networks , 2004, IEEE INFOCOM 2004.

[19]  Tom Kelly,et al.  Scalable TCP: improving performance in highspeed wide area networks , 2003, CCRV.

[20]  Rayadurgam Srikant,et al.  Diffusion approximations for models of congestion control in high-speed networks , 1998, Proceedings of the 37th IEEE Conference on Decision and Control (Cat. No.98CH36171).

[21]  Donald F. Towsley,et al.  Modeling TCP Reno performance: a simple model and its empirical validation , 2000, TNET.

[22]  Guido Appenzeller,et al.  Sizing router buffers , 2004, SIGCOMM '04.

[23]  Nick McKeown,et al.  Processor Sharing Flows in the Internet , 2005, IWQoS.

[24]  Mor Harchol-Balter,et al.  Analysis of SRPT scheduling: investigating unfairness , 2001, SIGMETRICS '01.