Date: Mon, 23 Jun 1997 22:40:57 -0400
From: "Pavel E. Guarisma" <peguaris@eos.ncsu.edu>
Subject: Q: a queue with a lognormal service distribution
To: Adam Wierzbicki <adamw@icm.edu.pl>
Hi!

An M/G/1 queueing model should work fine. You need to know the expected value and variance of your lognormal to use the M/G/1 model.

Also, you could try approximating the lognormal with a phase-type distribution. These distribution preserve Markovian properties and allow the use of well known models like M/M/*, M/H/*, M/E/*, etc.

Hope this helps,

Date: 23 Jun 1997 15:15:14 +0200
From: Adam Wierzbicki <adamw@icm.edu.pl>
Subject: Q: a queue with a lognormal service distribution
I am trying to analyse a system which has a lognormal service distribution (or some other, which resembles it -- it is skewed to the left) using a queuing model. Where can I find if work has been done on an analytical solution of such a system? Can someone suggest a way of working with it, if there are no known analytical solutions?

Date: Tue, 25 Feb 1997 13:31:51 -0500
From: llee@mit.edu
Subject: help with queueing problem
Hello,

I am not sure if this is the group to ask about queueing problems, but I hope someone can point me in the right direction.

The problem is as follows: I have a system with 2 servers A & B, which service customers with different exponentially distributed service times. The customer arrive into one queue with a Poisson distribution. The only twist is that there are 3 types of customers:

• A--which require the service of only A;
• B--which require the service of only B;
• AB--which require the service of either A OR B.

I'd like to know how to analyze this system for average waiting times, etc.

Is this a solved problem in queueing theory? If so, can someone tell me what it might be called in a standard textbook?

Many thanks!

Li Lee

Date: 26 Feb 1997 11:43:41 -0600
From: hennebry@plains.nodak.edu (Michael J. Hennebry)
Subject: help with queueing problem
The problem is not completely specified. What happens if neither server is busy and an AB customer arrives? Do the three kinds of customers all arrive with independant Poisson distributions?

Is the queue absotively FIFO? What happens if an A customer is in front, a B customer is next, and only server B is available?

If an AB customer arrives and neither server is busy, he picks the one with the faster service rate.

I didn't specify exactly how the queue works because I'm not so sure how to make the problem as tractable as possible. Ideally, I would say that each server, when it's done with its current job, goes down the queue to look for the first customer which can use its services. So if the queue is A, AB, B...., and server B is free, the AB customer gets served before A.

Thanks!

Date: Wed, 26 Feb 1997 18:21:16 -0500
From: Allan MacKinnon <allanmac@blueprint.com>
Subject: help with queueing problem
To: llee@mit.edu Hey, it's queueing theory -- chances are someone has already solved your problem! :)

You might want to look at the following paper:

Koenigsberg, E., Queueing With Special Service, Operations Research, vol 4., pp. 213-220, 1956(!)

I think it's a good match.

BTW, I came across this in Thomas Saaty's "Elements of Queueing Theory With Applications", 1961. It's a Dover book, so you can get it for $10. :)

ASM

Green, Linda, "A Queueing System with General-Use and Limited-Use Servers", _Operations_Research_, Vol. 33, No. 1 (January-February 1985), 168-182. Corrections in _Operations_Research_, Vol. 34, No. 1 (January-February 1986), 184.

Kaplan, Edward H., "A Public Housing Queue with Reneging and Task-Specific Servers", _Decision_Sciences_, Vol. 19, No. 2, (Spring 1988), 383-391.

D. A. Stanford and W. K. Grassmann

The Bilingual Server System: A Queueing Model Featuring Fully and Partially Qualified Servers. INFOR, vol 31, Nov. 1993.

Thank you.

• Henk C. Tijms, Stochastic Modelling and Analysis: A Computational Approach, Wiley, 1986.
• Ronald W. Wolff, Stochastic Modeling and the Theory of Queues, Prentice-Hall, 1989.
• Randolph Nelson, Probability, Stochastic Processes, and Queueing Theory, Springer-Verlag, 1995.

AUTHOR: Wolff, Ronald W., 1934- TITLE: Stochastic modeling and the theory of queues / IMPRINT: Englewood Cliffs, N.J. : Prentice Hall, c1989. CALL NUMBER: QA274.W65 1989

Meenan Vishnu

Date: 30 Jan 1997 21:54:22 GMT
From: grassman@cs.usask.ca (Dr. Grassmann)
Subject: Q: standard textbook on Queueing Theory
I am using Donald Gross and Carl M. Harris, Fundamentals of Queueing Theory by Wiley. I have the second edition, but I understand the third edition is coming out soon. Another well known book is by Leonard Kleinrock, Queueing Systems, Vol 1 and 2, John Wiley 1975.

Most frequently:

L. Kleinrock: Queueing Systems, Vols 1 and 2, Wiley, 1976.

Some others:

• R.B. Cooper, Introduction to Queueing Theory, North Holland, 1981.
• J. Walrand: An introduction to Queueing networks, Prentice Hall, 1988.
• D. Gross & C.M. Harris: Fundamentals of Queueing Theory, Wiley, 1974.
• H. C. Tijms, Stochastic Modelling and Analysis: A Computational Approach, Wiley, 1986.
• R.W. Wolff, Stochastic Modeling and the Theory of Queues, Prentice-Hall, 1989. Randolph Nelson, Probability, Stochastic Processes, and Queueing Theory, Springer-Verlag, 1995.

Thank you for the numerous reactions !!!!

Date: 17 Apr 1995 12:07:04 GMT
From: Yannis Tzavaras < jtza@intranet.gr >
Subject: Analysis of dynamic routing queues
I am interested to know if anyone is up to date with the status regarding the performance analysis of dynamic routing queues. The model of interest is the following and has been used to approximately model packet routing nodes:

Date: 21 Apr 1995 21:12:15 GMT
From: pong@bu.edu (Apinetr Unakul)
Subject: Mean Value Analysis
I have a problem getting my Load Dependent MVA to work properly. I don't know what went wrong with my MVA program and I just can't be sure what the right result should be. My program works for little example given in Jain's book but not another simple example given below.

This is a closed loop system with a total of 100 jobs (single class).

Thank you, Apinetr Unakul (Pong).

Date: 21 Sep 1995 11:38:15 -0400
From: shashank@kirk.ee.mcgill.ca (Shashank Nemawarkar)
Subject: Mean Value Analysis/Queuing Networks Refs Wanted
Hi,

I have questions regarding references on two problems in Queuing Networks, specifically the Mean Value Analysis (MVA).

The two problems are as follows:

1. For "simultaneous resource possession" problem: Let a service center C (say, processor) sends a customer (access) to the service center A (say, memory).However, the service center B (say, bus) should also be available during the service at A. Thus, B is also occupied/busy (simultaneously possessed) while A is providing the service.

   Note that in a multiprocessor system, A, B and C are subsystems of one processing node. A and B can service accesses from other processors. Similarly, C can send accesses to remote memory.

   Jacobson-Lazowska (Communications of the ACM,1982) propose a solution to above problem which requires two queuing network models to be solved iteratively.

   Q1==>Is there any paper, which uses only one queuing network model to solve this problem?

2. In multiple class queueing networks, different classes can have different service times at one or more service centers. The original MVA uses queue lengths at service centers to compute the waiting time for new customer. When the cost of service is high, and the number of visits of a by one class differs from those by other class, then error in performance prediction by original MVA can be significant.

   Q2==>Is there a reference which shows how to solve such a queuing network using MVA?

We would like to know any references in this regard.

Thank you,
Shashank

Date: 31 Oct 1995 05:42:50 GMT
From: jimkw@rmii.com (Jim Kwiecien)
Subject: Queuing Theory Book Recommendations
I am looking for a good book on both the theory and practical applications of Queuing Theory on a first graduate course level. Any recommedations?

Date: 1 Nov 1995 01:08:26 GMT
From: Mark Perkins <markp@robadome.com>
Subject: Queuing Theory Book Recommendations

I was quite happy with Gross, Donald, and Carl M. Harris, "Fundamentals of Queueing Theory," John Wiley & Sons, 1974. This text was used at Stanford for the first graduate course in queueing theory for both masters and PhD students. It seems to me to have a nice progression through the various queueing models with some theoritical foundations early in the book and then moving toward a more result-oriented presentation later in the book. Many examples of using the various calculations are also included.

Mark Perkins
Siemens Rolm Communications Inc.
markp@robadome.com

Date: 1 Nov 1995 15:55:42 GMT
From: mkgirish@engc.bu.edu (muckai girish)
Subject: Queuing Theory Book Recommendations
Jim Kwiecien (jimkw@rmii.com) wrote ...:

The most popular books are:

Muckai Krishnan Girish

Larry Russ
Stevens Institute of Technology
hoboken, NJ 07030

Date: 16 Nov 1995 04:04:43 GMT
From: engp4192@leonis.nus.sg (Hu Xuenian)
Subject: Help: Markov-modulated queue references needed
I am now working on a project related to Markov-modulated queueing system. Could someone kindly send me a reference list on this topic?

Thanks

Hu Xuenian
Dept. of Industrial & Systems Eng.
National Univ. of Singapore
Singapore

Date: Tue, 21 Nov 1995 03:18:33 GMT
From: YERKESRT%CS31@cadetmail.usafa.af.mil (RUSTIN T. YERKES)
Subject: M/Gamma/c Queueing Model
I am attempting to model a queue that has exponential interarrivals, a gamma service time distribution, and c=12 servers. I have been unable to locate any formulas or graphs that help solve the fundamental quantities of the queue. Any help is appreciated.

Rusty Yerkes, Cadet, USAFA
Operations Research Major
Email: YerkesRT96%cs31%usafa@cadetmail4.usafa.af.mil

Date: 8 Dec 1995 16:22:50 GMT
From: tankut@iastate.edu (Sabri T Atan)
Subject: M/Gamma/c Queueing Model

Tankut Atan
tankut@iastate.edu

"Achtung, baby!"

Excuse the noise caused by the rust falling off my brain cells - it's 30 years since I studied the subject:-).

Regards . . . . /Maurice Elliott.

Date: 22 Jan 1996 05:12:01 -0500
From: ilinm@aol.com (ILINM)
Subject: looking for queuing model with discrete service time
Hello,

I am looking for a reference to a queuing model that allows for a discretely distributed service time distribution. My service time consists of several discrete times that do not easily fit one of the more typical service time distributions (Exponential, Erlang, etc.). I have a poison arrival rate and multiple servers.

Thanks in advance,

Jeff
ILINM@oal.com

Date: 23 Jan 1996 12:20:23 GMT
From: kai.furmans@mach.uni-karlsruhe.de
Subject: looking for queuing model with discrete service time
In <4e090q$fq1@mira.sara.nl>, 9024794@edufee.fee.uva.nl writes ...:

If you need more information, please e-mail me.

Yours
Kai

Date: 26 Jan 1996 08:46:01 GMT
From: Erik Ljungberg <erik.ljungberg@promotor.telia.se>
Subject: Queuing simulator
I am looking for some kind of software were I can simulate different kind of queuing system and queuing algoritms. I need to measure factor as queuing time, owerflow and productivity. Is there anybody who have some ideas?

Erik Ljungberg
Telia Promotor AB
Uppsala, SWEDEN
Email: erik.ljungberg@promotor.telia.sweden

Date: Sat, 27 Jan 1996 02:25:48 GMT From: sullivan@indra.com (Steve Sullivan) Subject: Queuing simulator In article ...: Whew! There's a lot! See:

newsgroup: comp.simulation faq: ftp://piranha.eng.buffalo.edu/pub/comp.simulation/

Steve Sullivan

Date: Mon, 29 Jan 1996 21:21:27 GMT
From: dph@wind.bellcore.com (Daniel P Heyman)
Subject: looking for queuing model with discrete service time
In article <4e090q$fq1@mira.sara.nl> 9024794@edufee.fee.uva.nl writes:

Dan Heyman dph@bellcore.com

Date: 29 Jan 1996 14:26:07 GMT
From: Elia Barsoum <barsoum@cs.utwente.nl>
Subject: Queuing simulator
To: erik.ljungberg@promotor.telia.se

Dear Erik,

I think you can use SIMULA for such applications. I have used it several years ago to simulate a haven with ships,docks, storms and others! I measured waiting times, minimum number of small boats to draag the big ships. I don`t know if you can use SIMULA on a PC!.

Good luck..

Elia Barsoum

Date: 24 Jun 1996 13:34:30 GMT
From: labed@math-info.univ-paris5.fr (A.Labed)
Subject: MULTI CLASSES NETWORKS
I m looking for the performance parameters ( response time, throuput, mean queue length, ... ) for the analytical model given in the article "open, closed and mixed networks of queus with different classes of customers " Basket, Chandy, Muntz, Palacios ( BCMP)

In other words: performance parameters in multi classes queueing networks.

A. Labed.

Date: 28 Jun 1996 17:23:53 GMT
From: dtate+@pitt.edu (David M. Tate) Subject: Inverse Erlang-B Numerical Approximation
I am looking for a numerical subroutine to calculate approximate *inverse* Erlang loss formulae. That is, given a blocking rate and number of servers, I would like to (quickly) calculate the range of traffic intensities (or even a single qualifying traffic intensity) that would result in that blocking rate, given that many servers.

I already know how to do this by line search, using the forward recursive calculation of the number of servers required for a given traffic intensity and maximum blocking rate. I'm looking for a *fast* approximation, perhaps using rational functions, to give me a (closely) approximate answer without iteration.

Any references or suggestions will be appreciated. Thanks.

David Tate

Date: Thu, 04 Jul 1996 15:42:05 +0000
From: Shane Naughton <shane.naughton@tcd.ie>
Subject: Erlang-B/Erlang loss function properties
Hi,

I have a question about the properties of the Erlang loss function.

Given a set of input processes, each with independent poissonian arrival rates, but identical holding time distributions, and each with the same desired loss probability, is the average of the calculated servers to meet each individual loss probability equal to the calculated number of servers to meet the loss probability of a stream whose arrival rate is the average of the arrival rates of the set of streams?

i.e. given

Any help appreciated,

Thanks.

Date: 2 Jul 1996 22:35:59 -0400
From: wrandall@freenet.vcu.edu
Subject: Non-Poisson Queueing Problem
To Op.Researchers:

I am an engineer investigating water collection in the third world. Specifically, the queues at water collection points are my interest (for now). There is sufficient reason to believe that the arrivals are not Poisson, but dependent in many instances. After collecting data from queues in Honduras, I have begun to further explore the mathematical models available. Although lost in much of the math, it is clear that the Poisson assumption is critical to most of the models. I would greatly appreciate the ear (screen) of someone more familiar with queueing and the math behind it, in hopes that they can illuminate me to the essentials, or the sources where I may drink deeply.

Thanks, Bill Randall

Date: Tue, 9 Jul 1996 08:38:48 GMT
From: chana@elaine.ee.und.ac.za (AMISH)
Subject: BCMP network with finite queues
I would like to know if anyone has analysed a BCMP network with finite queue sizes (i.e. each queue is a M/M/1/K).

References to journals, papers or books would be helpful.]

Thanks

Amish
chana@eng.und.ac.za

Date: 9 Jul 1996 01:08:24 GMT
From: Rodney Beard <R.Beard@mailbox.uq.edu.au>
Subject: Non-Poisson Queueing Problem
To: wrandall@freenet.vcu.edu
wrandall@freenet.vcu.edu wrote:

What is the evidence for arrivals not being Poisson distributed? What distribution do they appear to follow? Or is it simply a case of bulk arrivals and /or batch processing.

Rodney Beard
University of Queensland

Date: 11 Jul 1996 11:49:54 -0500
From: rudi3964@utdallas.edu
Subject: Non-Poisson Queueing Problem
wrandall@freenet.vcu.edu wrote: ....

My question is in which sense are the arrivals not Poisson. A Poisson process has independent, stationary increments.

Independent -- The arrivals in period A are not correlated with the arrivals in period B.

Stationary -- the expected number of arrivals in period A is the same as the expected number of arrivals in an equal-sized period B.

If the number of arrivals near dusk is greater than at noon, but the number between 6:00 and 6:15 is independent of the number between 6:15 and 6:30, then they are independent but not stationary, and you can use a "thinning process" to generate a non-stationary Poisson arrival function.

On the other hand, if there are only twenty people coming at dusk, so if 17 show up before 6:15, then only three can show up after, then the process is not independent, and should be modelled completely differently.

If each home sends a collector each day, and the number of homes is small, then the model should simply identify each collector and the stochastic element for each one is the time of arrival, rather than using a Poisson process identifying the time periods with the stochastic element being number of arrivals.

To put it another way, if total visits per day is constant, the model should not assume it's stochastic.

Non-stationary Poisson processes are discussed in many books on Poisson processes and queuing theory. If you need a new model, we need more info before we can point you to it.

Jay Rudin
University of Texas at Dallas

Date: Tue, 08 Oct 1996 00:35:15 -0700
From: kostas Kontesis <kokos@athena.compulink.gr>
Subject: REQ:Queueing Tables
Dear O.R. fellows,

I'm looking for Queueing tables and Graphs - after a long-long research on the Net, nothing has been found. If someone of you happens to have such tables (tables that give mean waiting time in queues), PLEASE contact me!!!

Desperatelly needing those tables :-(

my e-mail: kokos@athena.compulink.gr

THANKS!!!

Date: Thu, 24 Oct 1996 11:51:33 -0700
From: gekko <gekko@dds.nl>
Subject: a queuing network with non-preemptive priority
Hello,

I'm a senior student Operations Research from Holland. At the moment I'm working on my thesis and the subject is about a queuing network with non-preemptive priority. I've done quite some time in research about this subject but still I haven't find anything (books, articles etc.). If you have read or seen anything about this topic, please let me know on my email-address : swan@econ.vu.nl. Thank you!

Susan Wan

Date: Fri, 08 Nov 1996 12:02:43 +1000
From: Gaius CO CHAN <gchan@tiny.me.su.oz.au>
Subject: Closed Network Queueing
Hi, Anybody know any basic book for captioned topic? I'm also looking for any newsgroup for this area.

Any idea?

Date: Sun, 10 Nov 1996 13:54:01 -0800
From: kostas Kontesis <
kokos@athena.compulink.gr>
Subject: Search for a book
Dear fellows,

I'm looking for this book

"Finite Queueing Tables" - Peck, L.G. and R.N. Hazelwood (1958)

I only know that it has become out-of-print. If anyone of you has any information about the book (where or how can I find it) or the authors (e-mail addresses, Universities they belong etc) please reply to this message or send me an e-mail. My e-address is:

kokos@athena.compulink.gr

Thanks in advance.

Dr Kostas Kontesis
Civil Engineer
Transportation Engineer
PhD Operations Research

Date: 11 Nov 1996 08:35:11 +1100
From: ask611@leonard.anu.edu.au (Alden S Klovdahl)
Subject: Search for a book
i hope you find it.

this request reminded me of the previous discussion on comp.theory on 'publishing on the web'. implicit in most of these discussions, which invariably raise the question - "how do we insure it (the web page) be available in the future?", is the assumption that the media we use for publishing today are available well into the future.

i've got a list of a dozen out-of-print books that i would buy on sight, but they are no longer available. true, libraries sometimes have them, and i can always pop over to the library of congress - only about 36 hours door-to-door - to look at them (assuming they haven't been stolen from the library, or misplaced).

would we have any less of this 'out-of-print' problem if we all published on the web?

regards, al

Date: Thu, 14 Nov 1996 11:51:05 GMT
From: pjbk@cee.hw.ac.uk (Peter JB King)
Subject: Closed Network Queueing

Crawl back to front page.