On the Impact of TCP and Per-Flow Scheduling on Internet Performance

Internet performance is tightly related to the properties of TCP and UDP protocols, jointly responsible for the delivery of the great majority of Internet traffic. It is well understood how these protocols behave under FIFO queuing and what are the network congestion effects. However, no comprehensive analysis is available when flow-aware mechanisms such as per-flow scheduling and dropping policies are deployed. Previous simulation and experimental results leave a number of unanswered questions. In the paper, we tackle this issue by modeling via a set of fluid non-linear ODEs the instantaneous throughput and the buffer occupancy of $N$ long-lived TCP sources under three per-flow scheduling disciplines (Fair Queuing, Longest Queue First, Shortest Queue First) and with longest queue drop buffer management. We study the system evolution and analytically characterize the stationary regime: closed-form expressions are derived for the stationary throughput/sending rate and buffer occupancy which give a thorough understanding of short/long-term fairness for TCP traffic. Similarly, we provide the characterization of the loss rate experienced by UDP flows in presence of TCP traffic. As a result, the analysis allows to quantify benefits and drawbacks related to the deployment of flow-aware scheduling mechanisms in different networking contexts. The model accuracy is confirmed by a set of $ns2$ simulations and by the evaluation of the three scheduling disciplines in a real implementation in the Linux kernel.

[1]  Raj Jain,et al.  A Quantitative Measure Of Fairness And Discrimination For Resource Allocation In Shared Computer Systems , 1998, ArXiv.

[2]  Thomas Bonald,et al.  Self-Prioritization of Audio and Video Traffic , 2011, 2011 IEEE International Conference on Communications (ICC).

[3]  Thomas Voice,et al.  Stability and fairness of explicit congestion control with small buffers , 2008, CCRV.

[4]  Scott Shenker,et al.  Approximate fairness through differential dropping , 2003, CCRV.

[5]  E. Leonardi,et al.  On the behavior of optimal scheduling algorithms under TCP sources , 2006, 2006 International Zurich Seminar on Communications.

[6]  François Baccelli,et al.  AIMD, fairness and fractal scaling of TCP traffic , 2002, Proceedings.Twenty-First Annual Joint Conference of the IEEE Computer and Communications Societies.

[7]  Abdesselem Kortebi,et al.  Evaluating the number of active flows in a scheduler realizing fair statistical bandwidth sharing , 2005, SIGMETRICS '05.

[8]  Ananth Balashankar.  Software Defined Networking , 2022 .

[9]  Guillaume Urvoy-Keller.  Institut Eurécom Service Differentiation by Means of Packet Scheduling , .

[10]  Abdesselem Kortebi,et al.  Minimizing the overhead in implementing flow-aware networking , 2005, 2005 Symposium on Architectures for Networking and Communications Systems (ANCS).

[11]  Frank Kelly,et al.  Rate control for communication networks: shadow prices, proportional fairness and stability , 1998, J. Oper. Res. Soc..

[12]  Hossam S. Hassanein,et al.  A performance comparison of class-based scheduling algorithms in future UMTS access , 2005, PCCC 2005. 24th IEEE International Performance, Computing, and Communications Conference, 2005..

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

[14]  Giovanna Carofiglio,et al.  On the Impact of TCP and Per-Flow Scheduling on Internet Performance , 2012, IEEE/ACM Transactions on Networking.

[15]  Nick McKeown,et al.  Software defined mobile networks , 2009, MobiHoc '09.

[16]  Gaurav Raina,et al.  Local bifurcation analysis of some dual congestion control algorithms , 2005, IEEE Transactions on Automatic Control.

[17]  Robert Tappan Morris,et al.  Dynamics of random early detection , 1997, SIGCOMM '97.

[18]  Nick McKeown,et al.  A Starvation-free Algorithm For Achieving 100% Throughput in an Input- Queued Switch , 1996 .

[19]  George Varghese,et al.  Efficient fair queueing using deficit round-robin , 1996, TNET.

[20]  François Baccelli,et al.  Proxy Caching in Split TCP: Dynamics, Stability and Tail Asymptotics , 2008, IEEE INFOCOM 2008 - The 27th Conference on Computer Communications.

[21]  John Doyle,et al.  Analysis of nonlinear delay differential equation models of TCP/AQM protocols using sums of squares , 2004, 2004 43rd IEEE Conference on Decision and Control (CDC) (IEEE Cat. No.04CH37601).

[22]  T. V. Lakshman,et al.  Design considerations for supporting TCP with per-flow queueing , 1998, Proceedings. IEEE INFOCOM '98, the Conference on Computer Communications. Seventeenth Annual Joint Conference of the IEEE Computer and Communications Societies. Gateway to the 21st Century (Cat. No.98.

[23]  A. Kortebi,et al.  On the Scalability of Fair Queueing , 2004 .

[24]  A. Kortebi,et al.  Cross-protect: implicit service differentiation and admission control , 2004, 2004 Workshop on High Performance Switching and Routing, 2004. HPSR..

[25]  R. Srikant,et al.  Multi-Path TCP: A Joint Congestion Control and Routing Scheme to Exploit Path Diversity in the Internet , 2006, IEEE/ACM Transactions on Networking.

[26]  Marco Ajmone Marsan,et al.  Using partial differential equations to model TCP mice and elephants in large IP networks , 2004, IEEE INFOCOM 2004.

[27]  Eitan Altman,et al.  TCP in presence of bursty losses , 2000, SIGMETRICS '00.

[28]  Isaac Keslassy,et al.  Modeling the interactions of congestion control and switch scheduling , 2009, 2009 17th International Workshop on Quality of Service.

[29]  Laurent Massoulié,et al.  Bandwidth sharing: objectives and algorithms , 2002, TNET.

[30]  Vishal Misra,et al.  Fluid-based analysis of a network of AQM routers supporting TCP flows with an application to RED , 2000, SIGCOMM.

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

[32]  John V. Guttag,et al.  Time-based Fairness Improves Performance in Multi-Rate WLANs , 2004, USENIX Annual Technical Conference, General Track.