How to improve the performance of ATM multiplexers

We consider an ATM multiplexer with a finite or infinite buffer and stochastic output rate. There are independent, identical sources connected to this multiplexer which transmit fluid directly into the buffer. We show that the throughput of the multiplexer is increasing in the number of sources connected, where we scale the cell stream process in such a way that the mean and peak input rate stays unchanged. In the finite buffer case, the cell loss is decreasing in the number of sources connected. Hence more links improve the performance of ATM multiplexers. In the last part, we prove that correlations within cell stream processes have a negative effect on the performance of ATM multiplexers. These investigations provide easily computable lower bounds to performance measures.

[1]  Tomasz Rolski,et al.  A MONOTONICITY RESULT FOR THE WORKLOAD IN MARKOV-MODULATED QUEUES , 1998 .

[2]  Hong Chen,et al.  Hierarchical Modeling of Stochastic Networks, Part I: Fluid Models , 1994 .

[3]  U. Rieder,et al.  Comparison Results for Markov-Modulated Recursive Models , 1997, Probability in the Engineering and Informational Sciences.

[4]  D. Mitra.  Stochastic theory of a fluid model of producers and consumers coupled by a buffer , 1988, Advances in Applied Probability.

[5]  John A. Buzacott,et al.  Stochastic models of manufacturing systems , 1993 .

[6]  J. George Shanthikumar,et al.  Discrete storage processes and their Poisson flow and fluid flow approximations , 1996, Queueing Syst. Theory Appl..

[7]  San-qi Li,et al.  Performance Trade-Offs in an Integrated Voice/Data Services TDM System , 1988, Perform. Evaluation.

[8]  J. George Shanthikumar,et al.  Regularity of Stochastic Processes: A Theory Based on Directional Convexity , 1993 .

[9]  Keith W. Ross,et al.  Multiservice Loss Models for Broadband Telecommunication Networks , 1997 .

[10]  A.I. Elwalid,et al.  Statistical multiplexing with loss priorities in rate-based congestion control of high-speed networks , 1994, IEEE Trans. Commun..

[11]  Nicole Bäuerle,et al.  The advantage of small machines in a stochastic fluid production process , 1998, Math. Methods Oper. Res..

[12]  J. Harrison,et al.  Brownian motion and stochastic flow systems , 1986 .

[13]  Moshe Shaked,et al.  Stochastic orders and their applications , 1994 .

[14]  Michael Pinedo,et al.  Integration of Discrete-Time Correlated Markov Processes in a TDM System , 1990, Probability in the Engineering and Informational Sciences.

[15]  N. Bäuerle.  Inequalities for stochastic models via supermodular orderings , 1997 .