Discriminatory processor sharing queues with stationary ergodic service times and the performance of TCP in overload

In a recent paper, Bonald and Roberts (2001) [6] studied non-persistent TCP connections in transient overload conditions, under the assumption that all connections have the same round-trip times. In this paper our goal is to develop theoretical tools that will enable us to relax this assumption and obtain explicit expressions for the rate of growth of the number of connections at the system, the rate at which TCP connections leave the system, as well as the time needed for the completion of a connection. To that end, we model the system as a discriminatory processor sharing (DPS) system which we analyze under very mild assumptions on the probability distributions related to different classes of arrivals: we only assume that the arrival rates of connections exist, and that the amount of information transmitted during a connection of a given type forms a stationary ergodic sequence. We then proceed to obtain explicit expressions for the growth rate of the number of connections at the DPS system for several specific probability distributions. We check through simulations the applicability of our queueing results for modeling TCP connections sharing a bottleneck.

[1]  F. Baccelli,et al.  Palm Probabilities and Stationary Queues , 1987 .

[2]  Thomas Bonald.  A Score-Based Opportunistic Scheduler for Fading Radio Channels , 2004 .

[3]  S. Stidham,et al.  Sample-Path Analysis of Queueing Systems , 1998 .

[4]  Bara Kim,et al.  Sojourn time distribution in the M/M/1 queue with discriminatory processor-sharing , 2004, Perform. Evaluation.

[5]  Eitan Altman,et al.  DPS queues with stationary ergodic service times and the performance of TCP in overload , 2004, IEEE INFOCOM 2004.

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

[7]  Mostafa Haddani.  Étude de modèles probabilistes de réseaux de télécommunication , 2001 .

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

[9]  Isi Mitrani,et al.  Sharing a Processor Among Many Job Classes , 1980, JACM.

[10]  Hong Chen,et al.  Fluid approximations for a processor-sharing queue , 1997, Queueing Syst. Theory Appl..

[11]  Anja Feldmann,et al.  Fitting mixtures of exponentials to long-tail distributions to analyze network performance models , 1997, Proceedings of INFOCOM '97.

[12]  Patrick Brown,et al.  Simulation Study of TCP in Overload , 2006, Advanced Int'l Conference on Telecommunications and Int'l Conference on Internet and Web Applications and Services (AICT-ICIW'06).

[13]  S. Grishechkin.  Crump–Mode–Jagers Branching Processes As a Method of Investigating ${\rm M}/{\rm G}/1$ Systems with Processor Sharing , 1992 .

[14]  A. S. Yashkova.  A NOTE ON LIMIT THEOREM FOR OVERLOADED PROCESSOR SHARING QUEUE , 2003 .

[15]  R. Núñez Queija,et al.  Discriminatory processor sharing revisited , 2005, Proceedings IEEE 24th Annual Joint Conference of the IEEE Computer and Communications Societies..

[16]  Thomas Bonald,et al.  Performance modeling of elastic traffic in overload , 2001, SIGMETRICS '01.

[17]  S. F. Yashkov,et al.  Processor sharing: A survey of the mathematical theory , 2007 .

[18]  Sem C. Borst,et al.  Asymptotic regimes and approximations for discriminatory processor sharing , 2004, PERV.

[19]  Alain Jean-Marie,et al.  On the transient behavior of the processor sharing queue , 1994, Queueing Syst. Theory Appl..

[20]  Alexander L. Stolyar,et al.  The Fluid Limit of an Overloaded Processor Sharing Queue , 2006, Math. Oper. Res..

[21]  Alan Weiss,et al.  A closed network with a discriminatory processor-sharing server , 1989, SIGMETRICS '89.

[22]  Konstantin Avrachenkov,et al.  Differentiation between short and long TCP flows: predictability of the response time , 2004, IEEE INFOCOM 2004.