Call packing bound for overflow loss systems

Finite loss queues with overflow naturally arise in a variety of communications structures. For these systems, there is no simple analytic expression for the loss probability. This paper proves and promotes easily computable bounds based on the so-called call packing principle. Under call packing, a standard product form expression is available. It is proven that call packing leads to a guaranteed upper bound for the loss probability. In addition, an analytic error bound for the accuracy is derived. This also leads to a secure lower bound. The call packing bound is also proven to be superior to the standard loss bound. Numerical results seem to indicate that the call packing bound is a substantial improvement over the standard loss bound and a quite reasonable upper bound approximation. The results seem to support a practical usefulness.

[1]  D. Sonderman Comparing multi-server queues with finite waiting rooms, II: Different numbers of servers , 1979, Advances in Applied Probability.

[2]  YONGJIAN ZHAO,et al.  Analysis on Partial Overflow Queueing Systems with Two Kinds of Calls , 1987, IEEE Trans. Commun..

[3]  Peter G. Taylor,et al.  Modeling Handovers in Cellular Mobile Networks with Dynamic Channel Allocation , 1995, Oper. Res..

[4]  Onno Boxma,et al.  ERMR: A generalised Equivalent Random Method for overflow systems with Repacking , 1998 .

[5]  R. I. Wilkinson The interconnection of telephone systems — Graded multiples , 1931 .

[6]  Peter G. Taylor,et al.  ALTERNATIVE ROUTING NETWORKS AND INTERRUPTIONS , 1988 .

[7]  R. I. Wilkinson Theories for toll traffic engineering in the U. S. A. , 1956 .

[8]  John F. Shortle An Equivalent Random Method with hyper-exponential service , 2004, Perform. Evaluation.

[9]  W. D. Ray,et al.  Stochastic Models: An Algorithmic Approach , 1995 .

[10]  N. M. van Dijk,et al.  A Note on Bounds and Error Bounds for nonexponential Batch Arrival Systems. , 1996 .

[11]  K. Sladký,et al.  A note on uniformization for dynamic non-negative systems , 2000 .

[12]  H. Akimaru,et al.  An Approximate Formula for Individual Call Losses in Overflow Systems , 1983, IEEE Trans. Commun..

[13]  Roch Guerin,et al.  Overflow analysis for finite waiting room systems , 1990, IEEE Trans. Commun..

[14]  Raymond A. Marie,et al.  Traffic conditioner: upper bound for the spacer overflow probability , 2004, Perform. Evaluation.

[15]  Ignas G. Niemegeers,et al.  Performability Modelling Tools and Techniques , 1996, Perform. Evaluation.

[16]  William A. Massey,et al.  New stochastic orderings for Markov processes on partially ordered spaces , 1984, The 23rd IEEE Conference on Decision and Control.

[17]  Bernard F. Lamond,et al.  Simple Bounds for Finite Single-Server Exponential Tandem Queues , 1988, Oper. Res..

[18]  Peter G. Taylor,et al.  A PRODUCT-FORM 'LOSS NETWORK' WITH A FORM OF QUEUEING , 1997 .

[19]  Andreas Brandt,et al.  Approximation for overflow moments of a multiservice link with trunk reservation , 2001, Perform. Evaluation.

[20]  L. Donatiello,et al.  On Evaluating the Cumulative Performance Distribution of Fault-Tolerant Computer Systems , 1991, IEEE Trans. Computers.

[21]  R. Schassberger The insensitivity of stationary probabilities in networks of queues , 1978, Advances in Applied Probability.

[22]  Jiyeon Lee,et al.  Asymptotics of overflow probabilities in Jackson networks , 2004, Oper. Res. Lett..

[23]  Youngsong Mun,et al.  Analysis of call packing and rearrangement in a multi stage switch , 1994, IEEE Trans. Commun..

[24]  A. Hordijk,et al.  Stochastic inequalities for an overflow model , 1987, Journal of Applied Probability.

[25]  Edmundo de Souza e Silva,et al.  Calculating transient distributions of cumulative reward , 1995, SIGMETRICS '95/PERFORMANCE '95.

[26]  A. Barbour Networks of queues and the method of stages , 1976, Advances in Applied Probability.

[27]  Richard R. Muntz,et al.  Bounding errors introduced by clustering of customers in closed product-form queuing networks , 1996, JACM.

[28]  S. Zachary,et al.  Loss networks , 2009, 0903.0640.

[29]  Donald F. Towsley,et al.  Exponential bounds with applications to call admission , 1997, JACM.

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

[31]  Muhammad El-Taha,et al.  Traffic overflow in loss systems with selective trunk reservation , 2000, Perform. Evaluation.

[32]  Peter G. Taylor,et al.  Strong Stochastic Bounds for the Stationary Distribution of a Class of Multicomponent Performability Models , 1998, Oper. Res..

[33]  Dirk P. Kroese,et al.  Efficient Estimation of Overflow Probabilities in Queues with Breakdowns , 1998, Perform. Evaluation.

[34]  Nico M. van Dijk,et al.  Bounds and error bounds for queueing networks , 1998, Ann. Oper. Res..

[35]  Erik A. van Doorn,et al.  On the Overflow Process from a Finite Markovian Queue , 1984, Perform. Evaluation.

[36]  Derek L. Eager,et al.  Bound hierarchies for multiple-class queuing networks , 1986, JACM.

[37]  Ger Koole,et al.  Approximating multi-skill blocking systems by HyperExponential Decomposition , 2006, Perform. Evaluation.

[38]  Ward Whitt,et al.  Comparison methods for queues and other stochastic models , 1986 .

[39]  Kurt Geihs,et al.  Bounds on buffer overflow probabilities in communication systems , 1982, Perform. Evaluation.