Permission is granted to make and distribute verbatim copies of this manual provided the copyright notice and this permission notice are preserved on all copies.

Permission is granted to copy and distribute modified versions of this manual under the conditions for verbatim copying, provided that the entire resulting derived work is distributed under the terms of a permission notice identical to this one.

Permission is granted to copy and distribute translations of this manual into another language, under the above conditions for modified versions.

queueing

This manual documents how to install and run the Queueing Toolbox. It corresponds to version 1.2.1 of the package.

• Summary
• Installation and Getting Started: Installation of the queueing toolbox.
• Markov Chains: Functions for Markov Chains.
• Single Station Queueing Systems: Functions for single-station queueing systems.
• Queueing Networks: Functions for queueing networks.
• References: References
• Copying: The GNU General Public License.
• Concept Index: An item for each concept.
• Function Index: An item for each function.
• Author Index: An item for each author.

1 Summary

• About the Queueing Toolbox: What is the Queueing Toolbox
• Contributing Guidelines: How to contribute
• Acknowledgements

1.1 About the Queueing Toolbox

This document describes the queueing toolbox for GNU Octave (queueing in short). The queueing toolbox, previously known as qnetworks, is a collection of functions written in GNU Octave for analyzing queueing networks and Markov chains. Specifically, queueing contains functions for analyzing Jackson networks, open, closed or mixed product-form BCMP networks, and computation of performance bounds. The following algorithms have been implemented

• Convolution for closed, single-class product-form networks with load-dependent service centers;
• Exact and approximate Mean Value Analysis (MVA) for single and multiple class product-form closed networks;
• MVA for mixed, multiple class product-form networks with load-independent service centers;
• Approximate MVA for closed, single-class networks with blocking (MVABLO algorithm by F. Akyildiz);
• Asymptotic Bounds, Balanced System Bounds and Geometric Bounds;

queueing provides functions for analyzing the following kind of single-station queueing systems:

• M/M/1
• M/M/m
• M/M/\infty
• M/M/1/k single-server, finite capacity system
• M/M/m/k multiple-server, finite capacity system
• Asymmetric M/M/m
• M/G/1 (general service time distribution)
• M/H_m/1 (Hyperexponential service time distribution)

Functions for Markov chain analysis are also provided:

• Birth-death process;
• Transient and steady-state occupancy probabilities;
• Mean times to absorption;
• Expected sojourn times and time-averaged sojourn times;
• Mean first passage times;

The queueing toolbox is distributed under the terms of the GNU General Public License (GPL), version 3 or later (see Copying). You are encouraged to share this software with others, and make this package more useful by contributing additional functions and reporting problems. See Contributing Guidelines.

If you use the queueing toolbox in a technical paper, please cite it as:

Moreno Marzolla, The qnetworks Toolbox: A Software Package for Queueing Networks Analysis. Khalid Al-Begain, Dieter Fiems and William J. Knottenbelt, Editors, Proceedings 17th International Conference on Analytical and Stochastic Modeling Techniques and Applications (ASMTA 2010) Cardiff, UK, June 14–16, 2010, volume 6148 of Lecture Notes in Computer Science, Springer, pp. 102–116, ISBN 978-3-642-13567-5

If you use BibTeX, this is the citation block:

@inproceedings{queueing,
  author    = {Moreno Marzolla},
  title     = {The qnetworks Toolbox: A Software Package for Queueing 
               Networks Analysis},
  booktitle = {Analytical and Stochastic Modeling Techniques and 
               Applications, 17th International Conference, 
               ASMTA 2010, Cardiff, UK, June 14-16, 2010. Proceedings},
  editor    = {Khalid Al-Begain and Dieter Fiems and William J. Knottenbelt},
  year      = {2010},
  publisher = {Springer},
  series    = {Lecture Notes in Computer Science},
  volume    = {6148},
  pages     = {102--116},
  ee        = {http://dx.doi.org/10.1007/978-3-642-13568-2_8},
  isbn      = {978-3-642-13567-5}
}

An early draft of the paper above is available as Technical Report UBLCS-2010-04, February 2010, Department of Computer Science, University of Bologna, Italy.

1.2 Contributing Guidelines

Contributions and bug reports are always welcome. If you want to contribute to the queueing package, here are some guidelines:

• If you are contributing a new function, please embed proper documentation within the function itself. The documentation must be in texinfo format, so that it can be extracted and formatted into the printable manual. See the existing functions of the queueing package for the documentation style.
• Make sure that each new function properly checks the validity of its input parameters. For example, each function accepting vectors should check whether the dimensions match.
• Provide bibliographic references for each new algorithm you contribute. If your implementation differs in some way from the reference you give, please describe how and why your implementation differs. Add references to the doc/references.txi file.
• Include test and demo blocks with your code. Test blocks are particularly important, since most algorithms tend to be quite tricky to implement correctly. If appropriate, test blocks should also verify that the function fails on incorrect input parameters.

Send your contribution to Moreno Marzolla (marzolla@cs.unibo.it). If you are just a user of this package and find it useful, let me know by dropping me a line. Thanks.

1.3 Acknowledgements

The following people (listed in alphabetical order) contributed to the queueing package, either by providing feedback, reporting bugs or contributing code: Philip Carinhas, Phil Colbourn, Diego Didona, Yves Durand, Marco Guazzone, Michele Mazzucco, Dmitry Kolesnikov.

2 Installation and Getting Started

• Installation through Octave package management system
• Manual installation
• Development sources
• Naming Conventions
• Quickstart Guide

2.1 Installation through Octave package management system

The most recent version of queueing is 1.2.1 and can be downloaded from Octave-Forge

http://octave.sourceforge.net/queueing/

Additional information can be found at

http://www.moreno.marzolla.name/software/queueing/

To install queueing, follow these steps:

1. If you have a recent version of GNU Octave and a network connection, you can install queueing directly from Octave command prompt using this command:

             octave:1> pkg install -forge queueing
   

   The command above will automaticall download and install the latest version of the queueing toolbox from Octave Forge, and install it on your machine.

   If you do not have root access, you can do a local install using:

             octave:1> pkg install -local -forge queueing
   

   This will install queueing within your home directory, and the package will be available to your user only.

2. Alternatively, you can first download queueing from Octave-Forge; then, to install the package in the system-wide location issue this command at the Octave prompt:

             octave:1> pkg install queueing-1.2.1.tar.gz
   

   (you may need to start Octave as root in order to allow the installation to copy the files to the target locations). After this, all functions will be readily available each time Octave starts, without the need to tweak the search path.

   If you do not have root access, you can do a local install using:

             octave:1> pkg install -local queueing-1.2.1.tar.gz
   

   Note: Octave version 3.2.3 as shipped with Ubuntu 10.04 LTS seems to ignore -local and always tries to install the package on the system directory.

3. Verify that the package is indeed installed using the pkg list command at the Octave prompt; after succesfull installation you should see something like that:

             octave:1>pkg list queueing
             Package Name  | Version | Installation directory
             --------------+---------+-----------------------
                 queueing  |   1.2.1 | /home/moreno/octave/queueing-1.2.1
   

4. Starting from version 1.1.1, queueing is no longer automatically loaded on Octave startup. To make the functions available for use, you need to issue the command

             octave:1>pkg load queueing
   

   at the Octave prompt. To automatically load queueing each time Octave starts, you can add the command above to the startup script (usually, ~/.octaverc on Unix systems).

5. To completely remove queueing from your system, use the pkg uninstall command:

             octave:1> pkg uninstall queueing
   

2.2 Manual installation

If you want to manually install queueing in a custom location, you can download the tarball and unpack it somewhere:

     tar xvfz queueing-1.2.1.tar.gz
     cd queueing-1.2.1/queueing/

Copy all .m files from the inst/ directory to some target location. Then, start Octave with the -p option to add the target location to the search path, so that Octave will find all queueing functions automatically:

     octave -p /path/to/queueing

For example, if all queueing m-files are in /usr/local/queueing, you can start Octave as follows:

     octave -p /usr/local/queueing

If you want, you can add the following line to ~/.octaverc:

     addpath("/path/to/queueing");

so that the path /path/to/queueing is automatically added to the search path each time Octave is started, and you no longer need to specify the -p option on the command line.

2.3 Development sources

The source code of the queueing package can be found in the Subversion repository at the URL:

http://octave.svn.sourceforge.net/viewvc/octave/trunk/octave-forge/main/queueing/

The source distribution contains additional development files which are not present in the installation tarball. This section briefly describes the content of the source tree. This is only relevant for developers who want to modify the code or documentation; normal users of the queueing package don't need

The source distribution contains the following directories:

doc/

Documentation source. Most of the documentation is extracted from the comment blocks of individual function files from the inst/ directory.

inst/

This directory contains the m-files which implement the various Queueing Network algorithms provided by queueing. As a notational convention, the names of source files containing functions for Queueing Networks start with the ‘qn’ prefix; the name of source files containing functions for Continuous-Time Markov Chains (CTMSs) start with the ‘ctmc’ prefix, and the names of files containing functions for Discrete-Time Markov Chains (DTMCs) start with the ‘dtmc’ prefix.

test/

This directory contains the test functions used to invoke all tests on all function files.

devel/

This directory contains function files which are either not working properly, or need additional testing before they are moved to the inst/ directory.

The queueing package ships with a Makefile which can be used to produce the documentation (in PDF and HTML format), and automatically execute all function tests. Specifically, the following targets are defined:

all

Running ‘make’ (or ‘make all’) on the top-level directory builds the programs used to extract the documentation from the comments embedded in the m-files, and then produce the documentation in PDF and HTML format (doc/queueing.pdf and doc/queueing.html, respectively).

check

Running ‘make check’ will execute all tests contained in the m-files. If you modify the code of any function in the inst/ directory, you should run the tests to ensure that no errors have been introduced. You are also encouraged to contribute new tests, especially for functions which are not adequately validated.

clean

distclean

dist

The ‘make clean’, ‘make distclean’ and ‘make dist’ commands are used to clean up the source directory and prepare the distribution archive in compressed tar format.

2.4 Naming Conventions

Most of the functions in the queueing package obey a common naming convention. Function names are made of several parts; the first part is a prefix which indicates the class of problems the function addresses:

ctmc-

Functions for continuous-time Markov chains

dtmc-

Functions for discrete-time Markov chains

qs-

Functions for analyzing queueing systems (individual service centers)

qn-

Functions for analyzing queueing networks

Functions dealing with Markov chains start with either the ctmc or dtmc prefix; the prefix is optionally followed by an additional string which hints at what the function does:

-bd

Birth-Death process

-mtta

Mean Time to Absorption

-fpt

First Passage Times

-exps

Expected Sojourn Times

-taexps

Time-Averaged Expected Sojourn Times

For example, function ctmcbd returns the infinitesimal generator matrix for a continuous birth-death process, while dtmcbd returns the transition probability matrix for a discrete birth-death process. Note that there exist functions ctmc and dtmc (without any suffix) that compute steady-state and transient state occupancy probabilities for CTMCs and DTMCs, respectively. See Markov Chains.

Functions whose name starts with qs- deal with single station queueing systems. The suffix describes the type of system, e.g., qsmm1 for M/M/1, qnmmm for M/M/m and so on. See Single Station Queueing Systems.

Finally, functions whose name starts with qn- deal with queueing networks. The character that follows indicates whether the function handles open ('o') or closed ('c') networks, and whether there is a single customer class ('s') or multiple classes ('m'). The string mix indicates that the function supports mixed networks with both open and closed customer classes.

-os-

Open, single-class network: open network with a single class of customers

-om-

Open, multiclass network: open network with multiple job classes

-cs-

Closed, single-class network

-cm-

Closed, multiclass network

-mix-

Mixed network with open and closed classes of customers

The last part of the function name indicates the algorithm implemented by the function. See Queueing Networks.

-aba

Asymptotic Bounds Analysis

-bsb

Balanced System Bounds

-gb

Geometric Bounds

-pb

PB Bounds

-cb

Composite Bounds (CB)

-mva

Mean Value Analysis (MVA) algorithm

-cmva

Conditional MVA

-mvald

MVA with general load-dependent servers

-mvaap

Approximate MVA

-mvablo

MVABLO approximation for blocking queueing networks

-conv

Convolution algorithm

-convld

Convolution algorithm with general load-dependent servers

The current version (1.2.1) of the queueing package still supports the old function names (although they are no longer documented and will disappear in future releases). However, calling one of the deprecated functions results in a warning message being displayed; the message appears only one time per session:

     octave:1> qnclosedab(10,[1 2 3])
         -| warning: qnclosedab is deprecated. Please use qncsaba instead
         ⇒ ans =  0.16667

Therefore, your legacy code should run unmodified with the current version of the queueing package. You can turn off all warning messages with the following command:

     octave:1> warning ("off", "qn:deprecated-function");

However, it is recommended to update to the new API and not ignore the warnings above. To help you catch usages of deprecated functions, even with applications which produce a lot of console output, you can transform warnings into errors so that your application will stop immediately:

     octave:1> warning ("error", "qn:deprecated-function");

2.5 Quickstart Guide

You can use all functions by simply invoking their name with the appropriate parameters; the queueing package should display an error message in case of missing/wrong parameters. Extensive documentation is provided for each function, and can be displayed with the help command. For example:

     octave:2> help qncsmvablo

prints the documentation for the qncsmvablo function. Additional information can be found in the queueing manual, which is available in PDF format in doc/queueing.pdf and in HTML format in doc/queueing.html.

Many functions have demo blocks containing code snippets which illustrate how that function can be used. For example, to execute the demos for the qnclosed function, use the demo command as follows:

     octave:4> demo qnclosed

Here we illustrate some basic usage of the queueing package by considering a few examples.

Example 1 Compute the stationary state occupancy probabilities of a continuous-time Markov chain with infinitesimal generator matrix

         / -0.8   0.6   0.2 \
     Q = |  0.3  -0.7   0.4 |
         \  0.2   0.2  -0.4 /

     Q = [ -0.8  0.6  0.2; \
            0.3 -0.7  0.4; \
            0.2  0.2 -0.4 ];
     q = ctmc(Q)
         ⇒ q = 0.23256   0.32558   0.44186

Example 2 Compute the transient state occupancy probability after n=3 transitions of a three state discrete time birth-death process, with birth probabilities \lambda_01 = 0.3 and \lambda_12 = 0.5 and death probabilities \mu_10 = 0.5 and \mu_21 = 0.7, assuming that the system is initially in state zero (i.e., the initial state occupancy probabilities are (1, 0, 0)).

     n = 3;
     p0 = [1 0 0];
     P = dtmcbd( [0.3 0.5], [0.5 0.7] );
     p = dtmc(P,n,p0)
         ⇒ p = 0.55300   0.29700   0.15000

Example 3 Compute server utilization, response time, mean number of requests and throughput of a closed queueing network with N=4 requests and three M/M/1–FCFS queues with mean service times \bf S = (1.0, 0.8, 1.4) and average number of visits \bf V = (1.0, 0.8, 0.8)

     S = [1.0 0.8 1.4];
     V = [1.0 0.8 0.8];
     N = 4;
     [U R Q X] = qncsmva(N, S, V)
         ⇒
          U = 0.70064   0.44841   0.78471
          R = 2.1030    1.2642    3.2433
          Q = 1.47346   0.70862   1.81792
          X = 0.70064   0.56051   0.56051

Example 4 Compute server utilization, response time, mean number of requests and throughput of an open queueing network with three M/M/1–FCFS queues with mean service times \bf S = (1.0, 0.8, 1.4) and average number of visits \bf V = (1.0, 0.8, 0.8). The overall arrival rate is \lambda = 0.8 req/s

     S = [1.0 0.8 1.4];
     V = [1.0 0.8 0.8];
     lambda = 0.8;
     [U R Q X] = qnos(lambda, S, V)
         ⇒
          U = 0.80000   0.51200   0.89600
          R = 5.0000    1.6393   13.4615
          Q = 4.0000    1.0492    8.6154
          X = 0.80000   0.64000   0.64000

3 Markov Chains

• Discrete-Time Markov Chains
• Continuous-Time Markov Chains

3.1 Discrete-Time Markov Chains

Let X_0, X_1, ..., X_n, ... be a sequence of random variables defined over a discete state space 1, 2, .... The sequence X_0, X_1, ..., X_n, ... is a stochastic process with discrete time 0, 1, 2, .... A Markov chain is a stochastic process {X_n, n=0, 1, 2, ...} which satisfies the following Markov property:

P(X_n+1 = x_n+1 | X_n = x_n, X_n-1 = x_n-1, ..., X_0 = x_0) = P(X_n+1 = x_n+1 | X_n = x_n)

which basically means that the probability that the system is in a particular state at time n+1 only depends on the state the system was at time n.

The evolution of a Markov chain with finite state space {1, 2, ..., N} can be fully described by a stochastic matrix \bf P(n) = [ P_i,j(n) ] such that P_i, j(n) = P( X_n+1 = j\ |\ X_n = i ). If the Markov chain is homogeneous (that is, the transition probability matrix \bf P(n) is time-independent), we can write \bf P = [P_i, j], where P_i, j = P( X_n+1 = j\ |\ X_n = i ) for all n=0, 1, ....

The transition probability matrix \bf P must satisfy the following two properties: (1) P_i, j ≥ 0 for all i, j, and (2) \sum_j=1^N P_i,j = 1 for all i

• State occupancy probabilities (DTMC)
• Birth-death process (DTMC)
• Expected number of visits (DTMC)
• Time-averaged expected sojourn times (DTMC)
• Mean time to absorption (DTMC)
• First passage times (DTMC)

3.1.1 State occupancy probabilities

Given a discrete-time Markov chain with state spate {1, 2, ..., N}, we denote with \bf \pi(n) = \left(\pi_1(n), \pi_2(n), ..., \pi_N(n) \right) the state occupancy probability vector at step n, n = 0, 1, .... \pi_i(n) denotes the probability that the system is in state i after n transitions.

Given the transition probability matrix \bf P and the initial state occupancy probability vector \bf \pi(0) = \left(\pi_1(0), \pi_2(0), ..., \pi_N(0)\right), \bf \pi(n) can be computed as:

     \pi(n) = \pi(0) P^n

Under certain conditions, there exists a stationary state occupancy probability \bf \pi = \lim_n \rightarrow +\infty \bf \pi(n), which is independent from \bf \pi(0). The stationary vector \bf \pi is the solution of the following linear system:

     /
     | \pi P   = \pi
     | \pi 1^T = 1
     \

where \bf 1 is the row vector of ones, and ( \cdot )^T the transpose operator.

EXAMPLE

This example is from GrSn97. Let us consider a maze with nine rooms, as shown in the following figure

     +-----+-----+-----+
     |     |     |     |
     |  1     2     3  |
     |     |     |     |
     +-   -+-   -+-   -+
     |     |     |     |
     |  4     5     6  |
     |     |     |     |
     +-   -+-   -+-   -+
     |     |     |     |
     |  7     8     9  |
     |     |     |     |
     +-----+-----+-----+

A mouse is placed in one of the rooms and can wander around. At each step, the mouse moves from the current room to a neighboring one with equal probability: if it is in room 1, it can move to room 2 and 4 with probability 1/2, respectively. If the mouse is in room 8, it can move to either 7, 5 or 9 with probability 1/3.

The transition probability \bf P from room i to room j is the following:

             / 0     1/2   0     1/2   0     0     0     0     0   \
             | 1/3   0     1/3   0     1/3   0     0     0     0   |
             | 0     1/2   0     0     0     1/2   0     0     0   |
             | 1/3   0     0     0     1/3   0     1/3   0     0   |
         P = | 0     1/4   0     1/4   0     1/4   0     1/4   0   |
             | 0     0     1/3   0     1/3   0     0     0     1/3 |
             | 0     0     0     1/2   0     0     0     1/2   0   |
             | 0     0     0     0     1/3   0     1/3   0     1/3 |
             \ 0     0     0     0     0     1/2   0     1/2   0   /

The stationary state occupancy probability vector can be computed using the following code:

      P = zeros(9,9);
      P(1,[2 4]    ) = 1/2;
      P(2,[1 5 3]  ) = 1/3;
      P(3,[2 6]    ) = 1/2;
      P(4,[1 5 7]  ) = 1/3;
      P(5,[2 4 6 8]) = 1/4;
      P(6,[3 5 9]  ) = 1/3;
      P(7,[4 8]    ) = 1/2;
      P(8,[7 5 9]  ) = 1/3;
      P(9,[6 8]    ) = 1/2;
      p = dtmc(P);
      disp(p)

         ⇒ 0.083333   0.125000   0.083333   0.125000
            0.166667   0.125000   0.083333   0.125000
            0.083333

3.1.2 Birth-death process

3.1.3 Expected Number of Visits

Given a N state discrete-time Markov chain with transition matrix \bf P and an integer n ≥ 0, we let L_i(n) be the the expected number of visits to state i during the first n transitions. The vector \bf L(n) = ( L_1(n), L_2(n), ..., L_N(n) ) is defined as

              n            n
             ___          ___
            \            \           i
     L(n) =  >   pi(i) =  >   pi(0) P
            /___         /___
             i=0          i=0

where \bf \pi(i) = \bf \pi(0)\bf P^i is the state occupancy probability after i transitions.

If \bf P is absorbing, i.e., the stochastic process eventually reaches a state with no outgoing transitions with probability 1, then we can compute the expected number of visits until absorption \bf L. To do so, we first rearrange the states to rewrite matrix \bf P as:

         / Q | R \
     P = |---+---|
         \ 0 | I /

where the first t states are transient and the last r states are absorbing (t+r = N). The matrix \bf N = (\bf I - \bf Q)^-1 is called the fundamental matrix; N_i,j is the expected number of times that the process is in the j-th transient state if it started in the i-th transient state. If we reshape \bf N to the size of \bf P (filling missing entries with zeros), we have that, for absorbing chains \bf L = \bf \pi(0)\bf N.

3.1.4 Time-averaged expected sojourn times

3.1.5 Mean Time to Absorption

The mean time to absorption is defined as the average number of transitions which are required to reach an absorbing state, starting from a transient state (or given an initial state occupancy probability vector \bf \pi(0)).

Let \bf t_i be the expected number of transitions before being absorbed in any absorbing state, starting from state i. Vector \bf t can be computed from the fundamental matrix \bf N (see Expected number of visits (DTMC)) as

     t = 1 N

Let \bf B = [ B_i, j ] be a matrix where B_i, j is the probability of being absorbed in state j, starting from transient state i. Again, using matrices \bf N and \bf R (see Expected number of visits (DTMC)) we can write

     B = N R

3.1.6 First Passage Times

The First Passage Time M_i, j is the average number of transitions needed to visit state j for the first time, starting from state i. Matrix \bf M satisfies the property that

                ___
               \
     M_ij = 1 + >   P_ij * M_kj
               /___
               k!=j

To compute \bf M = [ M_i, j] a different formulation is used. Let \bf W be the N \times N matrix having each row equal to the stationary state occupancy probability vector \bf \pi for \bf P; let \bf I be the N \times N identity matrix. Define \bf Z as follows:

                    -1
     Z = (I - P + W)

Then, we have that

            Z_jj - Z_ij
     M_ij = -----------
               \pi_j

According to the definition above, M_i,i = 0. We arbitrarily let M_i,i to be the mean recurrence time r_i for state i, that is the average number of transitions needed to return to state i starting from it. r_i is:

             1
     r_i = -----
           \pi_i

3.2 Continuous-Time Markov Chains

A stochastic process {X(t), t ≥ 0} is a continuous-time Markov chain if, for all integers n, and for any sequence t_0, t_1 , \ldots, t_n, t_n+1 such that t_0 < t_1 < \ldots < t_n < t_n+1, we have

P(X_n+1 = x_n+1 | X_n = x_n, X_n-1 = x_n-1, ..., X_0 = x_0) = P(X_n+1 = x_n+1 | X_n = x_n)

A continuous-time Markov chain is defined according to an infinitesimal generator matrix \bf Q = [Q_i,j], where for each i \neq j, Q_i, j is the transition rate from state i to state j. The matrix \bf Q must satisfy the property that, for all i, \sum_j=1^N Q_i, j = 0.

• State occupancy probabilities (CTMC)
• Birth-death process (CTMC)
• Expected sojourn times (CTMC)
• Time-averaged expected sojourn times (CTMC)
• Mean time to absorption (CTMC)
• First passage times (CTMC)

3.2.1 State occupancy probabilities

Similarly to the discrete case, we denote with \bf \pi(t) = (\pi_1(t), \pi_2(t), ..., \pi_N(t) ) the state occupancy probability vector at time t. \pi_i(t) is the probability that the system is in state i at time t ≥ 0.

Given the infinitesimal generator matrix \bf Q and the initial state occupancy probabilities \bf \pi(0) = (\pi_1(0), \pi_2(0), ..., \pi_N(0)), the state occupancy probabilities \bf \pi(t) at time t can be computed as:

     \pi(t) = \pi(0) exp(Qt)

where \exp( \bf Q t ) is the matrix exponential of \bf Q t. Under certain conditions, there exists a stationary state occupancy probability \bf \pi = \lim_t \rightarrow +\infty \bf \pi(t), which is independent from \bf \pi(0). \bf \pi is the solution of the following linear system:

     /
     | \pi Q   = 0
     | \pi 1^T = 1
     \

EXAMPLE

Consider a two-state CTMC such that transition rates between states are equal to 1. This can be solved as follows:

      Q = [ -1  1; \
             1 -1  ];
      q = ctmc(Q)

         ⇒ q = 0.50000   0.50000

3.2.2 Birth-Death Process

3.2.3 Expected Sojourn Times

Given a N state continuous-time Markov Chain with infinitesimal generator matrix \bf Q, we define the vector \bf L(t) = (L_1(t), L_2(t), \ldots, L_N(t)) such that L_i(t) is the expected sojourn time in state i during the interval [0,t), assuming that the initial occupancy probability at time 0 was \bf \pi(0). \bf L(t) can be expressed as the solution of the following differential equation:

      dL
      --(t) = L(t) Q + pi(0),    L(0) = 0
      dt

Alternatively, \bf L(t) can also be expressed in integral form as:

            / t
     L(t) = |   pi(u) du
            / 0

where \bf \pi(t) = \bf \pi(0) \exp(\bf Qt) is the state occupancy probability at time t; \exp(\bf Qt) is the matrix exponential of \bf Qt.

If there are absorbing states, we can define the vector expected sojourn times until absorption \bf L(\infty), where for each transient state i, L_i(\infty) is the expected total time spent in state i until absorption, assuming that the system started with a given state occupancy probability vector \bf \pi(0). Let \tau be the set of transient (i.e., non absorbing) states; let \bf Q_\tau be the restriction of \bf Q to the transient substates only. Similarly, let \bf \pi_\tau(0) be the restriction of the initial probability vector \bf \pi(0) to transient states \tau.

The expected time to absorption \bf L_\tau(\infty) is defined as the solution of the following equation:

     L_T( inf ) Q_T = -pi_T(0)

EXAMPLE

Let us consider a pure-birth, 4-states CTMC such that the transition rate from state i to state i+1 is \lambda_i = i \lambda (i=1, 2, 3), with \lambda = 0.5. The following code computes the expected sojourn time in state i, given the initial occupancy probability \bf \pi_0=(1,0,0,0).

      lambda = 0.5;
      N = 4;
      b = lambda*[1:N-1];
      d = zeros(size(b));
      Q = ctmcbd(b,d);
      t = linspace(0,10,100);
      p0 = zeros(1,N); p0(1)=1;
      L = zeros(length(t),N);
      for i=1:length(t)
        L(i,:) = ctmcexps(Q,t(i),p0);
      endfor
      plot( t, L(:,1), ";State 1;", "linewidth", 2, \
            t, L(:,2), ";State 2;", "linewidth", 2, \
            t, L(:,3), ";State 3;", "linewidth", 2, \
            t, L(:,4), ";State 4;", "linewidth", 2 );
      legend("location","northwest");
      xlabel("Time");
      ylabel("Expected sojourn time");

3.2.4 Time-Averaged Expected Sojourn Times

EXAMPLE

      lambda = 0.5;
      N = 4;
      birth = lambda*linspace(1,N-1,N-1);
      death = zeros(1,N-1);
      Q = diag(birth,1)+diag(death,-1);
      Q -= diag(sum(Q,2));
      t = linspace(1e-5,30,100);
      p = zeros(1,N); p(1)=1;
      M = zeros(length(t),N);
      for i=1:length(t)
        M(i,:) = ctmctaexps(Q,t(i),p);
      endfor
      clf;
      plot(t, M(:,1), ";State 1;", "linewidth", 2, \
           t, M(:,2), ";State 2;", "linewidth", 2, \
           t, M(:,3), ";State 3;", "linewidth", 2, \
           t, M(:,4), ";State 4 (absorbing);", "linewidth", 2 );
      legend("location","east");
      xlabel("Time");
      ylabel("Time-averaged Expected sojourn time");

3.2.5 Mean Time to Absorption

EXAMPLE

Let us consider a simple model of a redundant disk array. We assume that the array is made of 5 independent disks, such that the array can tolerate up to 2 disk failures without losing data. If three or more disks break, the array is dead and unrecoverable. We want to estimate the Mean-Time-To-Failure (MTTF) of the disk array.

We model this system as a 4 states Markov chain with state space \ 2, 3, 4, 5 \. State i denotes the fact that exactly i disks are active; state 2 is absorbing. Let \mu be the failure rate of a single disk. The system starts in state 5 (all disks are operational). We use a pure death process, with death rate from state i to state i-1 is \mu i, for i = 3, 4, 5).

The MTTF of the disk array is the MTTA of the Markov Chain, and can be computed with the following expression:

      mu = 0.01;
      death = [ 3 4 5 ] * mu;
      birth = 0*death;
      Q = ctmcbd(birth,death);
      t = ctmcmtta(Q,[0 0 0 1])

         ⇒ t = 78.333

REFERENCES

G. Bolch, S. Greiner, H. de Meer and K. Trivedi, Queueing Networks and Markov Chains: Modeling and Performance Evaluation with Computer Science Applications, Wiley, 1998.

3.2.6 First Passage Times

4 Single Station Queueing Systems

Single Station Queueing Systems contain a single station, and are thus quite easy to analyze. The queueing package contains functions for handling the following types of queues:

• The M/M/1 System: Single-server queueing station.
• The M/M/m System: Multiple-server queueing station.
• The M/M/inf System: Infinite-server (delay center) station.
• The M/M/1/K System: Single-server, finite-capacity queueing station.
• The M/M/m/K System: Multiple-server, finite-capacity queueing station.
• The Asymmetric M/M/m System: Asymmetric multiple-server queueing station.
• The M/G/1 System: Single-server with general service time distribution.
• The M/Hm/1 System: Single-server with hyperexponential service time distribution.

4.1 The M/M/1 System

The M/M/1 system is made of a single server connected to an unlimited FCFS queue. Requests arrive according to a Poisson process with rate \lambda; the service time is exponentially distributed with average service rate \mu. The system is stable if \lambda < \mu.

REFERENCES

G. Bolch, S. Greiner, H. de Meer and K. Trivedi, Queueing Networks and Markov Chains: Modeling and Performance Evaluation with Computer Science Applications, Wiley, 1998, Section 6.3.

4.2 The M/M/m System

The M/M/m system is similar to the M/M/1 system, except that there are m \geq 1 identical servers connected to a shared FCFS queue. Thus, at most m requests can be served at the same time. The M/M/m system can be seen as a single server with load-dependent service rate \mu(n), which is a function of the number n of requests in the system:

     mu(n) = min(m,n)*mu

where \mu is the service rate of each individual server.

REFERENCES

G. Bolch, S. Greiner, H. de Meer and K. Trivedi, Queueing Networks and Markov Chains: Modeling and Performance Evaluation with Computer Science Applications, Wiley, 1998, Section 6.5.

4.3 The M/M/inf System

The M/M/\infty system is similar to the M/M/m system, except that there are infinitely many identical servers (that is, m = \infty). Each new request is assigned to a new server, so that queueing never occurs. The M/M/\infty system is always stable.

REFERENCES

G. Bolch, S. Greiner, H. de Meer and K. Trivedi, Queueing Networks and Markov Chains: Modeling and Performance Evaluation with Computer Science Applications, Wiley, 1998, Section 6.4.

4.4 The M/M/1/K System

In a M/M/1/K finite capacity system there can be at most k jobs at any time. If a new request tries to join the system when there are already K other requests, the arriving request is lost. The queue has K-1 slots. The M/M/1/K system is always stable, regardless of the arrival and service rates \lambda and \mu.

4.5 The M/M/m/K System

The M/M/m/K finite capacity system is similar to the M/M/1/k system except that the number of servers is m, where 1 \leq m \leq K. The queue is made of K-m slots. The M/M/m/K system is always stable.

REFERENCES

G. Bolch, S. Greiner, H. de Meer and K. Trivedi, Queueing Networks and Markov Chains: Modeling and Performance Evaluation with Computer Science Applications, Wiley, 1998, Section 6.6.

4.6 The Asymmetric M/M/m System

The Asymmetric M/M/m system contains m servers connected to a single queue. Differently from the M/M/m system, in the asymmetric M/M/m each server may have a different service time.

REFERENCES

G. Bolch, S. Greiner, H. de Meer and K. Trivedi, Queueing Networks and Markov Chains: Modeling and Performance Evaluation with Computer Science Applications, Wiley, 1998

4.7 The M/G/1 System

4.8 The M/H_m/1 System

5 Queueing Networks

• Introduction to QNs: A brief introduction to Queueing Networks
• Single Class Models: Queueing Models with a single job class
• Multiple Class Models: Queueing Models with multiple job classess
• Generic Algorithms: High-level functions for QN analysis
• Bounds Analysis: Computation of asymptotic performance bounds
• QN Analysis Examples: Queueing Networks Analysis examples

5.1 Introduction to QNs

Queueing Networks (QN) are a very simple yet powerful modeling tool which can be used to analyze many kind of systems. In its simplest form, a QN is made of K service centers. Each center k has a queue, which is connected to m_k (generally identical) servers. Arriving customers (requests) join the queue if there is a slot available. Then, requests are served according to a (de)queueing policy (e.g., FIFO). After service completes, requests leave the server and can join another queue or exit from the system.

Service centers for which m_k = \infty are called delay centers or infinite servers. In this kind of centers, every request always finds one available server, so queueing never occurs.

Requests join the queue according to a queueing policy, such as:

FCFS

First-Come-First-Served

LCFS-PR

Last-Come-First-Served, Preemptive Resume

PS

Processor Sharing

IS

Infinite Server (for which m_k = \infty).

Queueing models can be open or closed. In open models there is an infinite population of requests; new customers are generated outside the system, and eventually leave the system. In closed systems there is a fixed population of request.

Queueing models can have a single request class (single class models), meaning that all requests behave in the same way (e.g., they spend the same average time on each particular server). In multiple class models there are multiple request classes, each one with its own parameters. Furthermore, in multiclass models there can be open and closed classes of requests within the same system.

A particular class of QN models, product-form QNs, is of particular interest. Product-form networks fulfill the following assumptions:

• The network can consist of open and closed job classes.
• The following queueing disciplines are allowed: FCFS, PS, LCFS-PR and IS.
• Service times for FCFS nodes must be exponentially distributed and class-independent. Service centers at PS, LCFS-PR and IS nodes can have any kind of service time distribution with a rational Laplace transform. Furthermore, for PS, LCFS-PR and IS nodes, different classes of customers can have different service times.
• The service rate of an FCFS node is only allowed to depend on the number of jobs at this node; in a PS, LCFS-PR and IS node the service rate for a particular job class can also depend on the number of jobs of that class at the node.
• In open networks two kinds of arrival processes are allowed: i) the arrival process is Poisson, with arrival rate \lambda which can depend on the number of jobs in the network. ii) the arrival process consists of U independent Poisson arrival streams where the U job sources are assigned to the U chains; the arrival rate can be load dependent.

Product-form networks are attractive because they be efficiently analyzed to compute steady-state performance measures.

5.2 Single Class Models

In single class models, all requests are indistinguishable and belong to the same class. This means that every request has the same average service time, and all requests move through the system with the same routing probabilities.

Model Inputs

lambda_k

(Open models only) External arrival rate to service center k.

lambda

(Open models only) Overall external arrival rate to the system as a whole: \lambda = \sum_k \lambda_k.

N

(Closed models only) Total number of requests in the system.

S_k

Average service time. S_k is the average service time on service center k. In other words, S_k is the average time from the instant in which a request is extracted from the queue and starts being service, and the instant at which service finishes and the request moves to another queue (or exits the system).

P_i, j

Routing probability matrix. \bf P = [P_i, j] is a K \times K matrix such that P_i, j is the probability that a request completing service at server i will move directly to server j, The probability that a request leaves the system after service at service center i is 1-\sum_j=1^K P_i, j.

V_k

Average number of visits. V_k is the average number of visits to the service center k. This quantity will be described shortly.

Model Outputs

U_k

Service center utilization. U_k is the utilization of service center k. The utilization is defined as the fraction of time in which the resource is busy (i.e., the server is processing requests).

R_k

Average response time. R_k is the average response time of service center k. The average response time is defined as the average time between the arrival of a customer in the queue, and the completion of service.

Q_k

Average number of customers. Q_k is the average number of requests in service center k. This includes both the requests in the queue, and the request being served.

X_k

Throughput. X_k is the throughput of service center k. The throughput is defined as the ratio of job completions (i.e., average number of jobs completed over a fixed interval of time).

Given these output parameters, additional performance measures can be computed as follows:

X

System throughput, X = X_1 / V_1

R

System response time, R = \sum_k=1^K R_k V_k

Q

Average number of requests in the system, Q = \sum_k=1 Q_k; for closed systems, this can be written as Q = N-XZ;

For open, single class models, the scalar \lambda denotes the external arrival rate of requests to the system. The average number of visits satisfy the following equation:

                       K
                      ___
                     \
     V_j = P_(0, j) + >   V_i P_(i, j)    j=1,...,K
                     /___
                      i=1

where P_0, j is the probability that an external arrival goes to service center j. If \lambda_j is the external arrival rate to service center j, and \lambda = \sum_j \lambda_j is the overall external arrival rate, then P_0, j = \lambda_j / \lambda.

For closed models, the visit ratios satisfy the following equation:

     /
     |         K
     |        ___
     |       \
     | V_j =  >   V_i P_(i, j)     j=1,...,K
     |       /___
     |        i=1
     |
     | V_r = 1                     for a selected reference station r
     \

Note that the set of traffic equations V_j = \sum_i=1^K V_i P_i, j alone can only be solved up to a multiplicative constant; to get a unique solution we impose an additional constraint V_r = 1. This constraint is equivalent to defining station r as the reference station (the default is r=1, See doc-qncsvisits). A job that returns to the reference station is assumed to have completed its activity cycle. The network throughput is set to the throughput of the reference station.

EXAMPLE

Figure 5.1: Closed network with a single class of requests

Figure 5.1 shows a closed QN with a single class of requests. The network has three service centers, labeled CPU, Disk1, Disk2. and represents the so called central server model of a computer system. Requests spend some time at the CPU, which is represented by a PS (Processor Sharing) node. After that, requests are routed to node Disk1 with probability 0.3, and to node Disk2 with probability 0.7. Both Disk1 and Disk2 are modeled as FCFS nodes.

If we enumerate the servers as CPU=1, Disk1=2, Disk2=3, we can define the routing matrix as follows:

         / 0  0.3  0.7 \
     P = | 1  0    0   |
         \ 1  0    0   /

The visit ratios V using station 1 as the reference station can be computed with:

      P = [0 0.3 0.7; \
           1 0   0  ; \
           1 0   0  ];
      V = qncsvisits(P)

        ⇒ V = 1.00000   0.30000   0.70000

EXAMPLE

Figure 5.2: Open Queueing Network with a single class of requests

Figure 5.2 shows a open QN with a single class of requests. The network has the same structure as the one in Figure 5.1, with the difference that here we have a stream of jobs arriving from outside the system, at a rate \lambda. After service completion at the CPU, a job can leave the system with probability 0.2, or be transferred to other nodes with the probabilities shown in the figure.

The routing matrix of this network is

         / 0  0.3  0.5 \
     P = | 1  0    0   |
         \ 1  0    0   /

If we let \lambda = 1.2, we can compute the visit ratios V as follows:

      p = 0.3;
      lambda = 1.2
      P = [0 0.3 0.5; \
           1 0   0  ; \
           1 0   0  ];
      V = qnosvisits(P,[1.2 0 0])

        ⇒ V = 5.0000   1.5000   2.5000

Function qnosvisits expects a vector with K elements as a second parameter, for open networks only. The vector contains the arrival rates at each individual node; since in our example external arrivals exist only for node S_1 with rate \lambda = 1.2, the second parameter is [1.2, 0, 0].

5.2.1 Open Networks

Jackson networks satisfy the following conditions:

• There is only one job class in the network; the overall number of jobs in the system is unlimited.
• There are N service centers in the network. Each service center may have Poisson arrivals from outside the system. A job can leave the system from any node.
• Arrival rates as well as routing probabilities are independent from the number of nodes in the network.
• External arrivals and service times at the service centers are exponentially distributed, and in general can be load-dependent.
• Service discipline at each node is FCFS

We define the joint probability vector \pi(k_1, k_2, \ldots, k_N) as the steady-state probability that there are k_i requests at service center i, for all i=1, 2, \ldots, N. Jackson networks have the property that the joint probability is the product of the marginal probabilities \pi_i:

     joint_prob = prod( pi )

where \pi_i(k_i) is the steady-state probability that there are k_i requests at service center i.

From the results computed by this function, it is possible to derive other quantities of interest as follows:

• System Response Time: The overall system response time can be computed as R_s = dot(V,R);
• Average number of requests: The average number of requests in the system can be computed as: Q_s = sum(Q)

EXAMPLE

      lambda = 3;
      V = [16 7 8];
      S = [0.01 0.02 0.03];
      [U R Q X] = qnos( lambda, S, V );
      R_s = dot(R,V) # System response time
      N = sum(Q) # Average number in system

     -| R_s =  1.4062
     -| N =  4.2186

REFERENCES

G. Bolch, S. Greiner, H. de Meer and K. Trivedi, Queueing Networks and Markov Chains: Modeling and Performance Evaluation with Computer Science Applications, Wiley, 1998.

5.2.2 Closed Networks

REFERENCES

M. Reiser and S. S. Lavenberg, Mean-Value Analysis of Closed Multichain Queuing Networks, Journal of the ACM, vol. 27, n. 2, April 1980, pp. 313–322. 10.1145/322186.322195

This implementation is described in R. Jain , The Art of Computer Systems Performance Analysis, Wiley, 1991, p. 577. Multi-server nodes are treated according to G. Bolch, S. Greiner, H. de Meer and K. Trivedi, Queueing Networks and Markov Chains: Modeling and Performance Evaluation with Computer Science Applications, Wiley, 1998, Section 8.2.1, "Single Class Queueing Networks".

EXAMPLE

      S = [ 0.125 0.3 0.2 ];
      V = [ 16 10 5 ];
      N = 20;
      m = ones(1,3);
      Z = 4;
      [U R Q X] = qncsmva(N,S,V,m,Z);
      X_s = X(1)/V(1); # System throughput
      R_s = dot(R,V); # System response time
      printf("\t    Util      Qlen     RespT      Tput\n");
      printf("\t--------  --------  --------  --------\n");
      for k=1:length(S)
        printf("Dev%d\t%8.4f  %8.4f  %8.4f  %8.4f\n", k, U(k), Q(k), R(k), X(k) );
      endfor
      printf("\nSystem\t          %8.4f  %8.4f  %8.4f\n\n", N-X_s*Z, R_s, X_s );

REFERENCES

M. Reiser and S. S. Lavenberg, Mean-Value Analysis of Closed Multichain Queuing Networks, Journal of the ACM, vol. 27, n. 2, April 1980, pp. 313–322. 10.1145/322186.322195

This implementation is described in G. Bolch, S. Greiner, H. de Meer and K. Trivedi, Queueing Networks and Markov Chains: Modeling and Performance Evaluation with Computer Science Applications, Wiley, 1998, Section 8.2.4.1, “Networks with Load-Deèpendent Service: Closed Networks”.

REFERENCES

G. Casale. A note on stable flow-equivalent aggregation in closed networks. Queueing Syst. Theory Appl., 60:193–-202, December 2008, 10.1007/s11134-008-9093-6

REFERENCES

This implementation is based on Edward D. Lazowska, John Zahorjan, G. Scott Graham, and Kenneth C. Sevcik, Quantitative System Performance: Computer System Analysis Using Queueing Network Models, Prentice Hall, 1984. http://www.cs.washington.edu/homes/lazowska/qsp/. In particular, see section 6.4.2.2 ("Approximate Solution Techniques").

According to the BCMP theorem, the state probability of a closed single class queueing network with K nodes and N requests can be expressed as:

     k = [k1, k2, ... kn]; population vector
     p = 1/G(N+1) \prod F(i,k);

Here \pi(k_1, k_2, \ldots, k_K) is the joint probability of having k_i requests at node i, for all i=1, 2, \ldots, K.

The convolution algorithms computes the normalization constants \bf G = \left(G(0), G(1), \ldots, G(N)\right) for single-class, closed networks with N requests. The normalization constants are returned as vector G=[G(1), G(2), ... G(N+1)] where G(i+1) is the value of G(i) (remember that Octave uses 1-base vectors). The normalization constant can be used to compute all performance measures of interest (utilization, average response time and so on).

queueing implements the convolution algorithm, in the function qncsconv and qncsconvld. The first one supports single-station nodes, multiple-station nodes and IS nodes. The second one supports networks with general load-dependent service centers.

NOTE

For a network with K service centers and N requests, this implementation of the convolution algorithm has time and space complexity O(NK).

REFERENCES

Jeffrey P. Buzen, Computational Algorithms for Closed Queueing Networks with Exponential Servers, Communications of the ACM, volume 16, number 9, september 1973, pp. 527–531. 10.1145/362342.362345

This implementation is based on G. Bolch, S. Greiner, H. de Meer and K. Trivedi, Queueing Networks and Markov Chains: Modeling and Performance Evaluation with Computer Science Applications, Wiley, 1998, pp. 313–317.

EXAMPLE

The normalization constant G can be used to compute the steady-state probabilities for a closed single class product-form Queueing Network with K nodes. Let k=[k_1, k_2, ..., k_K] be a valid population vector. Then, the steady-state probability p(i) to have k(i) requests at service center i can be computed as:

      k = [1 2 0];
      K = sum(k); # Total population size
      S = [ 1/0.8 1/0.6 1/0.4 ];
      m = [ 2 3 1 ];
      V = [ 1 .667 .2 ];
      [U R Q X G] = qncsconv( K, S, V, m );
      p = [0 0 0]; # initialize p
      # Compute the probability to have k(i) jobs at service center i
      for i=1:3
        p(i) = (V(i)*S(i))^k(i) / G(K+1) * \
               (G(K-k(i)+1) - V(i)*S(i)*G(K-k(i)) );
        printf("k(%d)=%d prob=%f\n", i, k(i), p(i) );
      endfor

     -| k(1)=1 prob=0.17975
     -| k(2)=2 prob=0.48404
     -| k(3)=0 prob=0.52779

REFERENCES

Herb Schwetman, Some Computational Aspects of Queueing Network Models, Technical Report CSD-TR-354, Department of Computer Sciences, Purdue University, feb, 1981 (revised).

M. Reiser, H. Kobayashi, On The Convolution Algorithm for Separable Queueing Networks, In Proceedings of the 1976 ACM SIGMETRICS Conference on Computer Performance Modeling Measurement and Evaluation (Cambridge, Massachusetts, United States, March 29–31, 1976). SIGMETRICS '76. ACM, New York, NY, pp. 109–117. 10.1145/800200.806187

This implementation is based on G. Bolch, S. Greiner, H. de Meer and K. Trivedi, Queueing Networks and Markov Chains: Modeling and Performance Evaluation with Computer Science Applications, Wiley, 1998, pp. 313–317. Function qncsconvld is slightly different from the version described in Bolch et al. because it supports general load-dependent centers (while the version in the book does not). The modification is in the definition of function F() in qncsconvld which has been made similar to function f_i defined in Schwetman, Some Computational Aspects of Queueing Network Models.

5.2.3 Non Product-Form QNs

REFERENCES

Ian F. Akyildiz, Mean Value Analysis for Blocking Queueing Networks, IEEE Transactions on Software Engineering, vol. 14, n. 2, april 1988, pp. 418–428. 10.1109/32.4663

5.3 Multiple Class Models

In multiple class QN models, we assume that there exist C different classes of requests. Each request from class c spends on average time S_c, k in service at service center k. For open models, we denote with \bf \lambda = \lambda_ck the arrival rates, where \lambda_c, k is the external arrival rate of class c customers at service center k. For closed models, we denote with \bf N = (N_1, N_2, \ldots, N_C) the population vector, where N_c is the number of class c requests in the system.

The transition probability matrix for these kind of networks will be a C \times K \times C \times K matrix \bf P = [P_r, i, s, j] such that P_r, i, s, j is the probability that a class r request which completes service at center i will join server j as a class s request.

Model input and outputs can be adjusted by adding additional indexes for the customer classes.

Model Inputs

lambda_{c, k}

External arrival rate of class-c requests to service center k

lambda

Overall external arrival rate to the whole system: \lambda = \sum_c \sum_k \lambda_c, k

S_c, k

Average service time. S_c, k is the average service time on service center k for class c requests.

P_r, i, s, j

Routing probability matrix. \bf P = [P_r, i, s, j] is a C \times K \times C \times K matrix such that P_r, i, s, j is the probability that a class r request which completes service at server i will move to server j as a class s request.

V_c, k

Average number of visits. V_c, k is the average number of visits of class c requests to center k.

Model Outputs

U_c, k

Utilization of service center k by class c requests. The utilization is defined as the fraction of time in which the resource is busy (i.e., the server is processing requests).

R_c, k

Average response time experienced by class c requests on service center k. The average response time is defined as the average time between the arrival of a customer in the queue, and the completion of service.

Q_c, k

Average number of class c requests on service center k. This includes both the requests in the queue, and the request being served.

X_c, k

Throughput of service center k for class c requests. The throughput is defined as the rate of completion of class c requests.

It is possible to define aggregate performance measures as follows:

U_k

Utilization of service center k: Uk = sum(U,k);

R_c

System response time for class c requests: Rc = sum( V.*R, 1 );

Q_c

Average number of class c requests in the system: Qc = sum( Q, 2 );

X_c

Class c throughput: Xc = X(:,1) ./ V(:,1);

We can define the visit ratios V_s, j for class s customers at service center j as follows:

V_sj = sum_r sum_i V_ri P_risj s=1,...,C, j=1,...,K V_s r_s = 1 s=1,...,C

where r_s is the class s reference station. Similarly to single class models, the traffic equation for closed multiclass networks can be solved up to multiplicative constants unless we choose one reference station for each closed chain class and set its visit ratio to 1.

For open networks the traffic equations are as follows:

V_sj = P_0sj + sum_r sum_i V_ri P_risj s=1,...,K, j=1,...,K

where P_0, s, j is the probability that an external arrival goes to service center j as a class-s request. If \lambda_s, j is the external arrival rate of class s requests to service center j, and \lambda = \sum_s \sum_j \lambda_s, j is the overall external arrival rate to the whole system, then P_0, s, j = \lambda_s, j / \lambda.

5.3.1 Open Networks

REFERENCES

Edward D. Lazowska, John Zahorjan, G. Scott Graham, and Kenneth C. Sevcik, Quantitative System Performance: Computer System Analysis Using Queueing Network Models, Prentice Hall, 1984. http://www.cs.washington.edu/homes/lazowska/qsp/. In particular, see section 7.4.1 ("Open Model Solution Techniques").

5.3.2 Closed Networks

REFERENCES

Herb Schwetman, Implementing the Mean Value Algorithm for the Solution of Queueing Network Models, Technical Report CSD-TR-355, Department of Computer Sciences, Purdue University, feb 15, 1982,

Note that the slightly different problem of generating all tuples k_1, k_2, \ldots, k_N such that \sum_i k_i = k and k_i are nonnegative integers, for some fixed integer k ≥ 0 has been described in S. Santini, Computing the Indices for a Complex Summation, unpublished report, available at http://arantxa.ii.uam.es/~ssantini/writing/notes/s668_summation.pdf

REFERENCES

Zahorjan, J. and Wong, E. The solution of separable queueing network models using mean value analysis. SIGMETRICS Perform. Eval. Rev. 10, 3 (Sep. 1981), 80-85. DOI 10.1145/1010629.805477

NOTE

Given a network with K service centers, C job classes and population vector \bf N=(N_1, N_2, \ldots, N_C), the MVA algorithm requires space O(C \prod_i (N_i + 1)). The time complexity is O(CK\prod_i (N_i + 1)). This implementation is slightly more space-efficient (see details in the code). While the space requirement can be mitigated by using some optimizations, the time complexity can not. If you need to analyze large closed networks you should consider the qncmmvaap function, which implements the approximate MVA algorithm. Note however that qncmmvaap will only provide approximate results.

REFERENCES

M. Reiser and S. S. Lavenberg, Mean-Value Analysis of Closed Multichain Queuing Networks, Journal of the ACM, vol. 27, n. 2, April 1980, pp. 313–322. 10.1145/322186.322195

This implementation is based on G. Bolch, S. Greiner, H. de Meer and K. Trivedi, Queueing Networks and Markov Chains: Modeling and Performance Evaluation with Computer Science Applications, Wiley, 1998 and Edward D. Lazowska, John Zahorjan, G. Scott Graham, and Kenneth C. Sevcik, Quantitative System Performance: Computer System Analysis Using Queueing Network Models, Prentice Hall, 1984. http://www.cs.washington.edu/homes/lazowska/qsp/. In particular, see section 7.4.2.1 ("Exact Solution Techniques").

REFERENCES

Y. Bard, Some Extensions to Multiclass Queueing Network Analysis, proc. 4th Int. Symp. on Modelling and Performance Evaluation of Computer Systems, feb. 1979, pp. 51–62.

P. Schweitzer, Approximate Analysis of Multiclass Closed Networks of Queues, Proc. Int. Conf. on Stochastic Control and Optimization, jun 1979, pp. 25–29.

This implementation is based on Edward D. Lazowska, John Zahorjan, G. Scott Graham, and Kenneth C. Sevcik, Quantitative System Performance: Computer System Analysis Using Queueing Network Models, Prentice Hall, 1984. http://www.cs.washington.edu/homes/lazowska/qsp/. In particular, see section 7.4.2.2 ("Approximate Solution Techniques"). This implementation is slightly different from the one described above, as it computes the average response times R instead of the residence times.

5.3.3 Mixed Networks

REFERENCES

Edward D. Lazowska, John Zahorjan, G. Scott Graham, and Kenneth C. Sevcik, Quantitative System Performance: Computer System Analysis Using Queueing Network Models, Prentice Hall, 1984. http://www.cs.washington.edu/homes/lazowska/qsp/. In particular, see section 7.4.3 ("Mixed Model Solution Techniques"). Note that in this function we compute the mean response time R instead of the mean residence time as in the reference.

Herb Schwetman, Implementing the Mean Value Algorithm for the Solution of Queueing Network Models, Technical Report CSD-TR-355, Department of Computer Sciences, Purdue University, feb 15, 1982,

5.4 Generic Algorithms

The queueing package provides a high-level function qnsolve for analyzing QN models. qnsolve takes as input a high-level description of the queueing model, and delegates the actual solution of the model to one of the lower-level function. qnsolve supports single or multiclass models, but at the moment only product-form networks can be analyzed. For non product-form networks See Non Product-Form QNs.

qnsolve accepts two input parameters. The first one is the list of nodes, encoded as an Octave cell array. The second parameter is the vector of visit ratios V, which can be either a vector (for single-class models) or a two-dimensional matrix (for multiple-class models).

Individual nodes in the network are structures build using the qnmknode function.

After the network has been defined, it is possible to solve it using qnsolve.

EXAMPLE

Let us consider a closed, multiclass network with C=2 classes and K=3 service center. Let the population be M=(2, 1) (class 1 has 2 requests, and class 2 has 1 request). The nodes are as follows:

• Node 1 is a M/M/1–FCFS node, with load-dependent service times. Service times are class-independent, and are defined by the matrix [0.2 0.1 0.1; 0.2 0.1 0.1]. Thus, S(1,2) = 0.2 means that service time for class 1 customers where there are 2 requests in 0.2. Note that service times are class-independent;
• Node 2 is a -/G/1–PS node, with service times S_1, 2 = 0.4 for class 1, and S_2, 2 = 0.6 for class 2 requests;
• Node 3 is a -/G/\infty node (delay center), with service times S_1, 3=1 and S_2, 3=2 for class 1 and 2 respectively.

After defining the per-class visit count V such that V(c,k) is the visit count of class c requests to service center k. We can define and solve the model as follows:

     
     
      QQ = { qnmknode( "m/m/m-fcfs", [0.2 0.1 0.1; 0.2 0.1 0.1] ), \
             qnmknode( "-/g/1-ps", [0.4; 0.6] ), \
             qnmknode( "-/g/inf", [1; 2] ) };
      V = [ 1 0.6 0.4; \
            1 0.3 0.7 ];
      N = [ 2 1 ];
      [U R Q X] = qnsolve( "closed", N, QQ, V );

EXAMPLE

      P = [0 0.3 0.7; 1 0 0; 1 0 0]; # Transition probability matrix
      S = [1 0.6 0.2];               # Average service times
      m = ones(size(S));             # All centers are single-server
      Z = 2;                         # External delay
      N = 15;                        # Maximum population to consider
      V = qncsvisits(P);             # Compute number of visits
      X_bsb_lower = X_bsb_upper = X_ab_lower = X_ab_upper = X_mva = zeros(1,N);
      for n=1:N
        [X_bsb_lower(n) X_bsb_upper(n)] = qncsbsb(n, S, V, m, Z);
        [X_ab_lower(n) X_ab_upper(n)] = qncsaba(n, S, V, m, Z);
        [U R Q X] = qnclosed( n, S, V, m, Z );
        X_mva(n) = X(1)/V(1);
      endfor
      close all;
      plot(1:N, X_ab_lower,"g;Asymptotic Bounds;", \
           1:N, X_bsb_lower,"k;Balanced System Bounds;", \
           1:N, X_mva,"b;MVA;", "linewidth", 2, \
           1:N, X_bsb_upper,"k", 1:N, X_ab_upper,"g" );
      axis([1,N,0,1]); legend("location","southeast");
      xlabel("Number of Requests n"); ylabel("System Throughput X(n)");

5.5 Bounds Analysis

REFERENCES

Edward D. Lazowska, John Zahorjan, G. Scott Graham, and Kenneth C. Sevcik, Quantitative System Performance: Computer System Analysis Using Queueing Network Models, Prentice Hall, 1984. http://www.cs.washington.edu/homes/lazowska/qsp/. In particular, see section 5.2 ("Asymptotic Bounds").

REFERENCES

Edward D. Lazowska, John Zahorjan, G. Scott Graham, and Kenneth C. Sevcik, Quantitative System Performance: Computer System Analysis Using Queueing Network Models, Prentice Hall, 1984. http://www.cs.washington.edu/homes/lazowska/qsp/. In particular, see section 5.4 ("Balanced Systems Bounds").

REFERENCES

Teemu Kerola, The Composite Bound Method (CBM) for Computing Throughput Bounds in Multiple Class Environments, Technical Report CSD-TR-475, Department of Computer Sciences, Purdue University, mar 13, 1984 (Revised aug 27, 1984).

REFERENCES

The original paper describing PB Bounds is C. H. Hsieh and S. Lam, Two classes of performance bounds for closed queueing networks, PEVA, vol. 7, n. 1, pp. 3–30, 1987

This function implements the non-iterative variant described in G. Casale, R. R. Muntz, G. Serazzi, Geometric Bounds: a Non-Iterative Analysis Technique for Closed Queueing Networks, IEEE Transactions on Computers, 57(6):780-794, June 2008.

REFERENCES

G. Casale, R. R. Muntz, G. Serazzi, Geometric Bounds: a Non-Iterative Analysis Technique for Closed Queueing Networks, IEEE Transactions on Computers, 57(6):780-794, June 2008. 10.1109/TC.2008.37

In this implementation we set X^+ and X^- as the upper and lower Asymptotic Bounds as computed by the qncsab function, respectively.

5.6 QN Analysis Examples

In this section we illustrate with a few examples how the queueing package can be used to evaluate queueing network models. Further examples can be found in the demo blocks of the functions described in this section, and can be accessed with the demo function Octave command.

5.6.1 Closed, Single Class Network

We now give a simple example on how the queueing toolbox can be used to analyze a closed network. Let us consider again the network shown in Figure 5.1. We denote with S_k the average service time at center k, k=1, 2, 3. We use S_1 = 1.0, S_2 = 2.0 and S_3 = 0.8. The routing of jobs within the network is described with a routing probability matrix \bf P. Specifically, a request completing service at center i is enqueued at center j with probability P_i, j. We use the following routing matrix:

         / 0  0.3  0.7 \
     P = | 1  0    0   |
         \ 1  0    0   /

The network above can be analyzed with the qnclosed function see doc-qnclosed. qnclosed requires the following parameters:

N

Number of requests in the network (since we are considering a closed network, the number of requests is fixed)

S

Array of average service times at the centers: S(k) is the average service time at center k.

V

Array of visit ratios: V(k) is the average number of visits to center k.

We can compute V_k from the routing probability matrix P_i, j using the qncsvisits function see doc-qncsvisits. We can analyze the network for a given population size N (for example, N=10) as follows:

     N = 10;
     S = [1 2 0.8];
     P = [0 0.3 0.7; 1 0 0; 1 0 0];
     V = qncsvisits(P);
     [U R Q X] = qnclosed( N, S, V )
        ⇒ U = 0.99139 0.59483 0.55518
        ⇒ R = 7.4360  4.7531  1.7500
        ⇒ Q = 7.3719  1.4136  1.2144
        ⇒ X = 0.99139 0.29742 0.69397

The output of qnclosed includes the vector of utilizations U_k at center k, response time R_k, average number of customers Q_k and throughput X_k. In our example, the throughput of center 1 is X_1 = 0.99139, and the average number of requests in center 3 is Q_3 = 1.2144. The utilization of center 1 is U_1 = 0.99139, which is the higher value among the service centers. Tus, center 1 is the bottleneck device.

This network can also be analyzed with the qnsolve function see doc-qnsolve. qnsolve can handle open, closed or mixed networks, and allows the network to be described in a very flexible way. First, let Q1, Q2 and Q3 be the variables describing the service centers. Each variable is instantiated with the qnmknode function.

     Q1 = qnmknode( "m/m/m-fcfs", 1 );
     Q2 = qnmknode( "m/m/m-fcfs", 2 );
     Q3 = qnmknode( "m/m/m-fcfs", 0.8 );

The first parameter of qnmknode is a string describing the type of the node. Here we use "m/m/m-fcfs" to denote a M/M/m–FCFS center. The second parameter gives the average service time. An optional third parameter can be used to specify the number m of service centers. If omitted, it is assumed m=1 (single-server node).

Now, the network can be analyzed as follows:

     N = 10;
     V = [1 0.3 0.7];
     [U R Q X] = qnsolve( "closed", N, { Q1, Q2, Q3 }, V )
        ⇒ U = 0.99139 0.59483 0.55518
        ⇒ R = 7.4360  4.7531  1.7500
        ⇒ Q = 7.3719  1.4136  1.2144
        ⇒ X = 0.99139 0.29742 0.69397

5.6.2 Open, Single Class Network

Open networks can be analyzed in a similar way. Let us consider an open network with K=3 service centers, and routing probability matrix as follows:

         / 0  0.3  0.5 \
     P = ! 1  0    0   |
         \ 1  0    0   /

In this network, requests can leave the system from center 1 with probability 1-(0.3+0.5) = 0.2. We suppose that external jobs arrive at center 1 with rate \lambda_1 = 0.15; there are no arrivals at centers 2 and 3.

Similarly to closed networks, we first need to compute the visit counts V_k to center k. We use the qnosvisits function as follows:

     P = [0 0.3 0.5; 1 0 0; 1 0 0];
     lambda = [0.15 0 0];
     V = qnosvisits(P, lambda)
        ⇒ V = 5.00000 1.50000 2.50000

where lambda(k) is the arrival rate at center k, and P is the routing matrix. Assuming the same service times as in the previous example, the network can be analyzed with the qnopen function see doc-qnopen, as follows:

     S = [1 2 0.8];
     [U R Q X] = qnopen( sum(lambda), S, V )
        ⇒ U = 0.75000 0.45000 0.30000
        ⇒ R = 4.0000  3.6364  1.1429
        ⇒ Q = 3.00000 0.81818 0.42857
        ⇒ X = 0.75000 0.22500 0.37500

The first parameter of the qnopen function is the (scalar) aggregate arrival rate.

Again, it is possible to use the qnsolve high-level function:

     Q1 = qnmknode( "m/m/m-fcfs", 1 );
     Q2 = qnmknode( "m/m/m-fcfs", 2 );
     Q3 = qnmknode( "m/m/m-fcfs", 0.8 );
     lambda = [0.15 0 0];
     [U R Q X] = qnsolve( "open", sum(lambda), { Q1, Q2, Q3 }, V )
        ⇒ U = 0.75000 0.45000 0.30000
        ⇒ R = 4.0000  3.6364  1.1429
        ⇒ Q = 3.00000 0.81818 0.42857
        ⇒ X = 0.75000 0.22500 0.37500

5.6.3 Closed Multiclass Network/1

The following example is taken from Herb Schwetman, Implementing the Mean Value Algorith for the Solution of Queueing Network Models, Technical Report CSD-TR-355, Department of Computer Sciences, Purdue University, feb 15, 1982.

We consider the following multiclass QN with three servers and two classes

Figure 5.3

Servers 1 and 2 (labeled APL and IMS, respectively) are infinite server nodes; server 3 (labeled SYS) is Processor Sharing (PS). Mean service times are given in the following table:

	APL	IMS	SYS
Class 1	1	-	0.025
Class 2	-	15	0.500

There is no class switching. If we assume a population of 15 requests for class 1, and 5 requests for class 2, then the model can be analyzed as follows:

      S = [1 0 .025; 0 15 .5];
      P = zeros(2,3,2,3);
      P(1,1,1,3) = P(1,3,1,1) = 1;
      P(2,2,2,3) = P(2,3,2,2) = 1;
      V = qncmvisits(P,[3 3]); # reference station is station 3
      N = [15 5];
      m = [-1 -1 1];
      [U R Q X] = qncmmva(N,S,V,m)

       ⇒
     U =
     
        14.32312    0.00000    0.35808
         0.00000    4.70699    0.15690
     
     R =
     
         1.00000    0.00000    0.04726
         0.00000   15.00000    0.93374
     
     Q =
     
        14.32312    0.00000    0.67688
         0.00000    4.70699    0.29301
     
     X =
     
        14.32312    0.00000   14.32312
         0.00000    0.31380    0.31380

5.6.4 Closed Multiclass Network/2

The following example is taken from M. Marzolla, The qnetworks Toolbox: A Software Package for Queueing Networks Analysis, Technical Report UBLCS-2010-04, Department of Computer Science, University of Bologna, Italy, February 2010.

Figure 5.4: Three-tier enterprise system model

The model shown in Figure 5.4 shows a three-tier enterprise system with K=6 service centers. The first tier contains the Web server (node 1), which is responsible for generating Web pages and transmitting them to clients. The application logic is implemented by nodes 2 and 3, and the storage tier is made of nodes 4–6.The system is subject to two workload classes, both represented as closed populations of N_1 and N_2 requests, respectively. Let D_c, k denote the service demand of class c requests at center k. We use the parameter values:

Serv. no.	Name	Class 1	Class 2
1	Web Server	12	2
2	App. Server 1	14	20
3	App. Server 2	23	14
4	DB Server 1	20	90
5	DB Server 2	80	30
6	DB Server 3	31	33

We set the total number of requests to 100, that is N_1 + N_2 = N = 100, and we study how different population mixes (N_1, N_2) affect the system throughput and response time. Let \beta_1 \in (0, 1) denote the fraction of class 1 requests: N_1 = \beta_1 N, N_2 = (1-\beta_1)N. The following Octave code defines the model for \beta_1 = 0.1:

     N = 100;     # total population size
     beta1 = 0.1; # fraction of class 1 reqs.
     S = [12 14 23 20 80 31; \
           2 20 14 90 30 33 ];
     V = ones(size(S));
     pop = [fix(beta1*N) N-fix(beta1*N)];
     [U R Q X] = qncmmva(pop, S, V);

The qncmmva(pop, S, V) function invocation (line 7) uses the multiclass MVA algorithm to compute per-class utilizations U_c, k, response times R_c,k, mean queue lengths Q_c,k and throughputs X_c,k at each service center k, given a population vector pop, mean service times S and visit ratios V. Since we are given the service demands D_c, k = S_c, k V_c,k, but function qncmmva() requires separate service times and visit ratios, we set the service times equal to the demands (line 3–4), and all visit ratios equal to one (line 5). Overall class and system throughputs and response times can also be computed:

     X1 = X(1,1) / V(1,1)     # class 1 throughput
             ⇒ X1 =  0.0044219
     X2 = X(2,1) / V(2,1)     # class 2 throughput
             ⇒ X2 =  0.010128
     XX = X1 + X2             # system throughput
             ⇒ XX =  0.014550
     R1 = dot(R(1,:), V(1,:)) # class 1 resp. time
             ⇒ R1 =  2261.5
     R2 = dot(R(2,:), V(2,:)) # class 2 resp. time
             ⇒ R2 =  8885.9
     RR = N / XX              # system resp. time
             ⇒ RR =  6872.7

dot(X,Y) computes the dot product of two vectors. R(1,:) is the first row of matrix R and V(1,:) is the first row of matrix V, so dot(R(1,:), V(1,:)) computes \sum_k R_1,k V_1,k.

Figure 5.5: Throughput and Response Times as a function of the population mix

We can also compute the system power \Phi = X / R, which defines how efficiently resources are being used: high values of \Phi denote the desirable situation of high throughput and low response time. fig:power shows \Phi as a function of \beta_1. We observe a “plateau” of the global system power, corresponding to values of \beta_1 which approximately lie between 0.3 and 0.7. The per-class power exhibits an interesting (although not completely surprising) pattern, where the class with higher population exhibits worst efficiency as it produces higher contention on the resources.

Figure 5.6: System Power as a function of the population mix

5.6.5 Closed Multiclass Network/3

We now consider an example of multiclass network with class switching. The example is taken from Sch82, and is shown in Figure fig:class_switching.

Figure 5.7: Multiclass Model with Class Switching

The system consists of three devices and two job classes. The CPU node is a PS server, while the two nodes labeled I/O are FCFS. Class 1 mean service time at the CPU is 0.01; class 2 mean service time at the CPU is 0.05. The mean service time at node 2 is 0.1, and is class-independent. Similarly, the mean service time at node 3 is 0.07. Jobs in class 1 leave the CPU and join class 2 with probability 0.1; jobs of class 2 leave the CPU and join class 1 with probability 0.2. There are N=3 jobs, which are initially allocated to class 1. However, note that since class switching is allowed, the total number of jobs in each class does not remain constant; however the total number of jobs does.

      C = 2; K = 3;
      S = [.01 .07 .10; \
           .05 .07 .10 ];
      P = zeros(C,K,C,K);
      P(1,1,1,2) = .7; P(1,1,1,3) = .2; P(1,1,2,1) = .1;
      P(2,1,2,2) = .3; P(2,1,2,3) = .5; P(2,1,1,1) = .2;
      P(1,2,1,1) = P(2,2,2,1) = 1;
      P(1,3,1,1) = P(2,3,2,1) = 1;
      N = [3 0];
      [U R Q X] = qncmmva(N, S, P)

       ⇒
     U =
     
        0.12609   0.61784   0.25218
        0.31522   0.13239   0.31522
     
     R =
     
        0.014653   0.133148   0.163256
        0.073266   0.133148   0.163256
     
     Q =
     
        0.18476   1.17519   0.41170
        0.46190   0.25183   0.51462
     
     X =
     
        12.6089    8.8262    2.5218
         6.3044    1.8913    3.1522

6 References

[Aky88]

Ian F. Akyildiz, Mean Value Analysis for Blocking Queueing Networks, IEEE Transactions on Software Engineering, vol. 14, n. 2, april 1988, pp. 418–428. DOI 10.1109/32.4663

[Bar79]

Y. Bard, Some Extensions to Multiclass Queueing Network Analysis, proc. 4th Int. Symp. on Modelling and Performance Evaluation of Computer Systems, feb. 1979, pp. 51–62.

[BCMP75]

F. Baskett, K. Mani Chandy, R. R. Muntz, and F. G. Palacios. 1975. Open, Closed, and Mixed Networks of Queues with Different Classes of Customers. J. ACM 22, 2 (April 1975), 248—260, DOI 10.1145/321879.321887

[BGMT98]

G. Bolch, S. Greiner, H. de Meer and K. Trivedi, Queueing Networks and Markov Chains: Modeling and Performance Evaluation with Computer Science Applications, Wiley, 1998.

[Buz73]

J. P. Buzen, Computational Algorithms for Closed Queueing Networks with Exponential Servers, Communications of the ACM, volume 16, number 9, september 1973, pp. 527–531. DOI 10.1145/362342.362345

[C08]

G. Casale, A note on stable flow-equivalent aggregation in closed networks. Queueing Syst. Theory Appl., 60:193–-202, December 2008, DOI 10.1007/s11134-008-9093-6

[CMS08]

G. Casale, R. R. Muntz, G. Serazzi, Geometric Bounds: a Non-Iterative Analysis Technique for Closed Queueing Networks, IEEE Transactions on Computers, 57(6):780-794, June 2008. DOI 10.1109/TC.2008.37

[GrSn97]

C. M. Grinstead, J. L. Snell, (July 1997). Introduction to Probability. American Mathematical Society. ISBN 978-0821807491; this excellent textbook is available in PDF format and can be used under the terms of the GNU Free Documentation License (FDL)

[Jac04]

J. R. Jackson, Jobshop-Like Queueing Systems, Vol. 50, No. 12, Ten Most Influential Titles of "Management Science's" First Fifty Years (Dec., 2004), pp. 1796-1802, available online

[Jai91]

R. Jain, The Art of Computer Systems Performance Analysis, Wiley, 1991, p. 577.

[HsLa87]

C. H. Hsieh and S. Lam, Two classes of performance bounds for closed queueing networks, PEVA, vol. 7, n. 1, pp. 3–30, 1987

[Ker84]

T. Kerola, The Composite Bound Method (CBM) for Computing Throughput Bounds in Multiple Class Environments, Technical Report CSD-TR-475, Department of Computer Sciences, Purdue University, mar 13, 1984 (Revised aug 27, 1984).

[LZGS84]

E. D. Lazowska, J. Zahorjan, G. Scott Graham, and K. C. Sevcik, Quantitative System Performance: Computer System Analysis Using Queueing Network Models, Prentice Hall, 1984. available online.

[ReKo76]

M. Reiser, H. Kobayashi, On The Convolution Algorithm for Separable Queueing Networks, In Proceedings of the 1976 ACM SIGMETRICS Conference on Computer Performance Modeling Measurement and Evaluation (Cambridge, Massachusetts, United States, March 29–31, 1976). SIGMETRICS '76. ACM, New York, NY, pp. 109–117. DOI 10.1145/800200.806187

[ReLa80]

M. Reiser and S. S. Lavenberg, Mean-Value Analysis of Closed Multichain Queuing Networks, Journal of the ACM, vol. 27, n. 2, April 1980, pp. 313–322. DOI 10.1145/322186.322195

[Sch79]

P. Schweitzer, Approximate Analysis of Multiclass Closed Networks of Queues, Proc. Int. Conf. on Stochastic Control and Optimization, jun 1979, pp. 25—29

[Sch81]

H. Schwetman, Some Computational Aspects of Queueing Network Models, Technical Report CSD-TR-354, Department of Computer Sciences, Purdue University, feb, 1981 (revised).

[Sch82]

H. Schwetman, Implementing the Mean Value Algorithm for the Solution of Queueing Network Models, Technical Report CSD-TR-355, Department of Computer Sciences, Purdue University, feb 15, 1982.

[Tij03]

H. C. Tijms, A first course in stochastic models, John Wiley and Sons, 2003, ISBN 0471498807, ISBN 9780471498803, DOI 10.1002/047001363X

[ZaWo81]

J. Zahorjan and E. Wong, The solution of separable queueing network models using mean value analysis. SIGMETRICS Perform. Eval. Rev. 10, 3 (Sep. 1981), 80-85. DOI DOI 10.1145/1010629.805477

Appendix A GNU GENERAL PUBLIC LICENSE

     Copyright © 2007 Free Software Foundation, Inc. http://fsf.org/
     
     Everyone is permitted to copy and distribute verbatim copies of this
     license document, but changing it is not allowed.

Preamble

The GNU General Public License is a free, copyleft license for software and other kinds of works.

The licenses for most software and other practical works are designed to take away your freedom to share and change the works. By contrast, the GNU General Public License is intended to guarantee your freedom to share and change all versions of a program—to make sure it remains free software for all its users. We, the Free Software Foundation, use the GNU General Public License for most of our software; it applies also to any other work released this way by its authors. You can apply it to your programs, too.

When we speak of free software, we are referring to freedom, not price. Our General Public Licenses are designed to make sure that you have the freedom to distribute copies of free software (and charge for them if you wish), that you receive source code or can get it if you want it, that you can change the software or use pieces of it in new free programs, and that you know you can do these things.

To protect your rights, we need to prevent others from denying you these rights or asking you to surrender the rights. Therefore, you have certain responsibilities if you distribute copies of the software, or if you modify it: responsibilities to respect the freedom of others.

For example, if you distribute copies of such a program, whether gratis or for a fee, you must pass on to the recipients the same freedoms that you received. You must make sure that they, too, receive or can get the source code. And you must show them these terms so they know their rights.

Developers that use the GNU GPL protect your rights with two steps: (1) assert copyright on the software, and (2) offer you this License giving you legal permission to copy, distribute and/or modify it.

For the developers' and authors' protection, the GPL clearly explains that there is no warranty for this free software. For both users' and authors' sake, the GPL requires that modified versions be marked as changed, so that their problems will not be attributed erroneously to authors of previous versions.

Some devices are designed to deny users access to install or run modified versions of the software inside them, although the manufacturer can do so. This is fundamentally incompatible with the aim of protecting users' freedom to change the software. The systematic pattern of such abuse occurs in the area of products for individuals to use, which is precisely where it is most unacceptable. Therefore, we have designed this version of the GPL to prohibit the practice for those products. If such problems arise substantially in other domains, we stand ready to extend this provision to those domains in future versions of the GPL, as needed to protect the freedom of users.

Finally, every program is threatened constantly by software patents. States should not allow patents to restrict development and use of software on general-purpose computers, but in those that do, we wish to avoid the special danger that patents applied to a free program could make it effectively proprietary. To prevent this, the GPL assures that patents cannot be used to render the program non-free.

The precise terms and conditions for copying, distribution and modification follow.

TERMS AND CONDITIONS

0. Definitions.

   “This License” refers to version 3 of the GNU General Public License.

   “Copyright” also means copyright-like laws that apply to other kinds of works, such as semiconductor masks.

   “The Program” refers to any copyrightable work licensed under this License. Each licensee is addressed as “you”. “Licensees” and “recipients” may be individuals or organizations.

   To “modify” a work means to copy from or adapt all or part of the work in a fashion requiring copyright permission, other than the making of an exact copy. The resulting work is called a “modified version” of the earlier work or a work “based on” the earlier work.

   A “covered work” means either the unmodified Program or a work based on the Program.

   To “propagate” a work means to do anything with it that, without permission, would make you directly or secondarily liable for infringement under applicable copyright law, except executing it on a computer or modifying a private copy. Propagation includes copying, distribution (with or without modification), making available to the public, and in some countries other activities as well.

   To “convey” a work means any kind of propagation that enables other parties to make or receive copies. Mere interaction with a user through a computer network, with no transfer of a copy, is not conveying.

   An interactive user interface displays “Appropriate Legal Notices” to the extent that it includes a convenient and prominently visible feature that (1) displays an appropriate copyright notice, and (2) tells the user that there is no warranty for the work (except to the extent that warranties are provided), that licensees may convey the work under this License, and how to view a copy of this License. If the interface presents a list of user commands or options, such as a menu, a prominent item in the list meets this criterion.

1. Source Code.

   The “source code” for a work means the preferred form of the work for making modifications to it. “Object code” means any non-source form of a work.

   A “Standard Interface” means an interface that either is an official standard defined by a recognized standards body, or, in the case of interfaces specified for a particular programming language, one that is widely used among developers working in that language.

   The “System Libraries” of an executable work include anything, other than the work as a whole, that (a) is included in the normal form of packaging a Major Component, but which is not part of that Major Component, and (b) serves only to enable use of the work with that Major Component, or to implement a Standard Interface for which an implementation is available to the public in source code form. A “Major Component”, in this context, means a major essential component (kernel, window system, and so on) of the specific operating system (if any) on which the executable work runs, or a compiler used to produce the work, or an object code interpreter used to run it.

   The “Corresponding Source” for a work in object code form means all the source code needed to generate, install, and (for an executable work) run the object code and to modify the work, including scripts to control those activities. However, it does not include the work's System Libraries, or general-purpose tools or generally available free programs which are used unmodified in performing those activities but which are not part of the work. For example, Corresponding Source includes interface definition files associated with source files for the work, and the source code for shared libraries and dynamically linked subprograms that the work is specifically designed to require, such as by intimate data communication or control flow between those subprograms and other parts of the work.

   The Corresponding Source need not include anything that users can regenerate automatically from other parts of the Corresponding Source.

   The Corresponding Source for a work in source code form is that same work.

2. Basic Permissions.

   All rights granted under this License are granted for the term of copyright on the Program, and are irrevocable provided the stated conditions are met. This License explicitly affirms your unlimited permission to run the unmodified Program. The output from running a covered work is covered by this License only if the output, given its content, constitutes a covered work. This License acknowledges your rights of fair use or other equivalent, as provided by copyright law.

   You may make, run and propagate covered works that you do not convey, without conditions so long as your license otherwise remains in force. You may convey covered works to others for the sole purpose of having them make modifications exclusively for you, or provide you with facilities for running those works, provided that you comply with the terms of this License in conveying all material for which you do not control copyright. Those thus making or running the covered works for you must do so exclusively on your behalf, under your direction and control, on terms that prohibit them from making any copies of your copyrighted material outside their relationship with you.

   Conveying under any other circumstances is permitted solely under the conditions stated below. Sublicensing is not allowed; section 10 makes it unnecessary.

3. Protecting Users' Legal Rights From Anti-Circumvention Law.

   No covered work shall be deemed part of an effective technological measure under any applicable law fulfilling obligations under article 11 of the WIPO copyright treaty adopted on 20 December 1996, or similar laws prohibiting or restricting circumvention of such measures.

   When you convey a covered work, you waive any legal power to forbid circumvention of technological measures to the extent such circumvention is effected by exercising rights under this License with respect to the covered work, and you disclaim any intention to limit operation or modification of the work as a means of enforcing, against the work's users, your or third parties' legal rights to forbid circumvention of technological measures.

4. Conveying Verbatim Copies.

   You may convey verbatim copies of the Program's source code as you receive it, in any medium, provided that you conspicuously and appropriately publish on each copy an appropriate copyright notice; keep intact all notices stating that this License and any non-permissive terms added in accord with section 7 apply to the code; keep intact all notices of the absence of any warranty; and give all recipients a copy of this License along with the Program.

   You may charge any price or no price for each copy that you convey, and you may offer support or warranty protection for a fee.

5. Conveying Modified Source Versions.

   You may convey a work based on the Program, or the modifications to produce it from the Program, in the form of source code under the terms of section 4, provided that you also meet all of these conditions:

   1. The work must carry prominent notices stating that you modified it, and giving a relevant date.
   2. The work must carry prominent notices stating that it is released under this License and any conditions added under section 7. This requirement modifies the requirement in section 4 to “keep intact all notices”.
   3. You must license the entire work, as a whole, under this License to anyone who comes into possession of a copy. This License will therefore apply, along with any applicable section 7 additional terms, to the whole of the work, and all its parts, regardless of how they are packaged. This License gives no permission to license the work in any other way, but it does not invalidate such permission if you have separately received it.
   4. If the work has interactive user interfaces, each must display Appropriate Legal Notices; however, if the Program has interactive interfaces that do not display Appropriate Legal Notices, your work need not make them do so.

   A compilation of a covered work with other separate and independent works, which are not by their nature extensions of the covered work, and which are not combined with it such as to form a larger program, in or on a volume of a storage or distribution medium, is called an “aggregate” if the compilation and its resulting copyright are not used to limit the access or legal rights of the compilation's users beyond what the individual works permit. Inclusion of a covered work in an aggregate does not cause this License to apply to the other parts of the aggregate.

6. Conveying Non-Source Forms.

   You may convey a covered work in object code form under the terms of sections 4 and 5, provided that you also convey the machine-readable Corresponding Source under the terms of this License, in one of these ways:

   1. Convey the object code in, or embodied in, a physical product (including a physical distribution medium), accompanied by the Corresponding Source fixed on a durable physical medium customarily used for software interchange.
   2. Convey the object code in, or embodied in, a physical product (including a physical distribution medium), accompanied by a written offer, valid for at least three years and valid for as long as you offer spare parts or customer support for that product model, to give anyone who possesses the object code either (1) a copy of the Corresponding Source for all the software in the product that is covered by this License, on a durable physical medium customarily used for software interchange, for a price no more than your reasonable cost of physically performing this conveying of source, or (2) access to copy the Corresponding Source from a network server at no charge.
   3. Convey individual copies of the object code with a copy of the written offer to provide the Corresponding Source. This alternative is allowed only occasionally and noncommercially, and only if you received the object code with such an offer, in accord with subsection 6b.
   4. Convey the object code by offering access from a designated place (gratis or for a charge), and offer equivalent access to the Corresponding Source in the same way through the same place at no further charge. You need not require recipients to copy the Corresponding Source along with the object code. If the place to copy the object code is a network server, the Corresponding Source may be on a different server (operated by you or a third party) that supports equivalent copying facilities, provided you maintain clear directions next to the object code saying where to find the Corresponding Source. Regardless of what server hosts the Corresponding Source, you remain obligated to ensure that it is available for as long as needed to satisfy these requirements.
   5. Convey the object code using peer-to-peer transmission, provided you inform other peers where the object code and Corresponding Source of the work are being offered to the general public at no charge under subsection 6d.

   A separable portion of the object code, whose source code is excluded from the Corresponding Source as a System Library, need not be included in conveying the object code work.

   A “User Product” is either (1) a “consumer product”, which means any tangible personal property which is normally used for personal, family, or household purposes, or (2) anything designed or sold for incorporation into a dwelling. In determining whether a product is a consumer product, doubtful cases shall be resolved in favor of coverage. For a particular product received by a particular user, “normally used” refers to a typical or common use of that class of product, regardless of the status of the particular user or of the way in which the particular user actually uses, or expects or is expected to use, the product. A product is a consumer product regardless of whether the product has substantial commercial, industrial or non-consumer uses, unless such uses represent the only significant mode of use of the product.

   “Installation Information” for a User Product means any methods, procedures, authorization keys, or other information required to install and execute modified versions of a covered work in that User Product from a modified version of its Corresponding Source. The information must suffice to ensure that the continued functioning of the modified object code is in no case prevented or interfered with solely because modification has been made.

   If you convey an object code work under this section in, or with, or specifically for use in, a User Product, and the conveying occurs as part of a transaction in which the right of possession and use of the User Product is transferred to the recipient in perpetuity or for a fixed term (regardless of how the transaction is characterized), the Corresponding Source conveyed under this section must be accompanied by the Installation Information. But this requirement does not apply if neither you nor any third party retains the ability to install modified object code on the User Product (for example, the work has been installed in ROM).

   The requirement to provide Installation Information does not include a requirement to continue to provide support service, warranty, or updates for a work that has been modified or installed by the recipient, or for the User Product in which it has been modified or installed. Access to a network may be denied when the modification itself materially and adversely affects the operation of the network or violates the rules and protocols for communication across the network.

   Corresponding Source conveyed, and Installation Information provided, in accord with this section must be in a format that is publicly documented (and with an implementation available to the public in source code form), and must require no special password or key for unpacking, reading or copying.

7. Additional Terms.

   “Additional permissions” are terms that supplement the terms of this License by making exceptions from one or more of its conditions. Additional permissions that are applicable to the entire Program shall be treated as though they were included in this License, to the extent that they are valid under applicable law. If additional permissions apply only to part of the Program, that part may be used separately under those permissions, but the entire Program remains governed by this License without regard to the additional permissions.

   When you convey a copy of a covered work, you may at your option remove any additional permissions from that copy, or from any part of it. (Additional permissions may be written to require their own removal in certain cases when you modify the work.) You may place additional permissions on material, added by you to a covered work, for which you have or can give appropriate copyright permission.

   Notwithstanding any other provision of this License, for material you add to a covered work, you may (if authorized by the copyright holders of that material) supplement the terms of this License with terms:

   1. Disclaiming warranty or limiting liability differently from the terms of sections 15 and 16 of this License; or
   2. Requiring preservation of specified reasonable legal notices or author attributions in that material or in the Appropriate Legal Notices displayed by works containing it; or
   3. Prohibiting misrepresentation of the origin of that material, or requiring that modified versions of such material be marked in reasonable ways as different from the original version; or
   4. Limiting the use for publicity purposes of names of licensors or authors of the material; or
   5. Declining to grant rights under trademark law for use of some trade names, trademarks, or service marks; or
   6. Requiring indemnification of licensors and authors of that material by anyone who conveys the material (or modified versions of it) with contractual assumptions of liability to the recipient, for any liability that these contractual assumptions directly impose on those licensors and authors.

   All other non-permissive additional terms are considered “further restrictions” within the meaning of section 10. If the Program as you received it, or any part of it, contains a notice stating that it is governed by this License along with a term that is a further restriction, you may remove that term. If a license document contains a further restriction but permits relicensing or conveying under this License, you may add to a covered work material governed by the terms of that license document, provided that the further restriction does not survive such relicensing or conveying.

   If you add terms to a covered work in accord with this section, you must place, in the relevant source files, a statement of the additional terms that apply to those files, or a notice indicating where to find the applicable terms.

   Additional terms, permissive or non-permissive, may be stated in the form of a separately written license, or stated as exceptions; the above requirements apply either way.

8. Termination.

   You may not propagate or modify a covered work except as expressly provided under this License. Any attempt otherwise to propagate or modify it is void, and will automatically terminate your rights under this License (including any patent licenses granted under the third paragraph of section 11).

   However, if you cease all violation of this License, then your license from a particular copyright holder is reinstated (a) provisionally, unless and until the copyright holder explicitly and finally terminates your license, and (b) permanently, if the copyright holder fails to notify you of the violation by some reasonable means prior to 60 days after the cessation.

   Moreover, your license from a particular copyright holder is reinstated permanently if the copyright holder notifies you of the violation by some reasonable means, this is the first time you have received notice of violation of this License (for any work) from that copyright holder, and you cure the violation prior to 30 days after your receipt of the notice.

   Termination of your rights under this section does not terminate the licenses of parties who have received copies or rights from you under this License. If your rights have been terminated and not permanently reinstated, you do not qualify to receive new licenses for the same material under section 10.

9. Acceptance Not Required for Having Copies.

   You are not required to accept this License in order to receive or run a copy of the Program. Ancillary propagation of a covered work occurring solely as a consequence of using peer-to-peer transmission to receive a copy likewise does not require acceptance. However, nothing other than this License grants you permission to propagate or modify any covered work. These actions infringe copyright if you do not accept this License. Therefore, by modifying or propagating a covered work, you indicate your acceptance of this License to do so.

10. Automatic Licensing of Downstream Recipients.

    Each time you convey a covered work, the recipient automatically receives a license from the original licensors, to run, modify and propagate that work, subject to this License. You are not responsible for enforcing compliance by third parties with this License.

    An “entity transaction” is a transaction transferring control of an organization, or substantially all assets of one, or subdividing an organization, or merging organizations. If propagation of a covered work results from an entity transaction, each party to that transaction who receives a copy of the work also receives whatever licenses to the work the party's predecessor in interest had or could give under the previous paragraph, plus a right to possession of the Corresponding Source of the work from the predecessor in interest, if the predecessor has it or can get it with reasonable efforts.

    You may not impose any further restrictions on the exercise of the rights granted or affirmed under this License. For example, you may not impose a license fee, royalty, or other charge for exercise of rights granted under this License, and you may not initiate litigation (including a cross-claim or counterclaim in a lawsuit) alleging that any patent claim is infringed by making, using, selling, offering for sale, or importing the Program or any portion of it.

11. Patents.

    A “contributor” is a copyright holder who authorizes use under this License of the Program or a work on which the Program is based. The work thus licensed is called the contributor's “contributor version”.

    A contributor's “essential patent claims” are all patent claims owned or controlled by the contributor, whether already acquired or hereafter acquired, that would be infringed by some manner, permitted by this License, of making, using, or selling its contributor version, but do not include claims that would be infringed only as a consequence of further modification of the contributor version. For purposes of this definition, “control” includes the right to grant patent sublicenses in a manner consistent with the requirements of this License.

    Each contributor grants you a non-exclusive, worldwide, royalty-free patent license under the contributor's essential patent claims, to make, use, sell, offer for sale, import and otherwise run, modify and propagate the contents of its contributor version.

    In the following three paragraphs, a “patent license” is any express agreement or commitment, however denominated, not to enforce a patent (such as an express permission to practice a patent or covenant not to sue for patent infringement). To “grant” such a patent license to a party means to make such an agreement or commitment not to enforce a patent against the party.

    If you convey a covered work, knowingly relying on a patent license, and the Corresponding Source of the work is not available for anyone to copy, free of charge and under the terms of this License, through a publicly available network server or other readily accessible means, then you must either (1) cause the Corresponding Source to be so available, or (2) arrange to deprive yourself of the benefit of the patent license for this particular work, or (3) arrange, in a manner consistent with the requirements of this License, to extend the patent license to downstream recipients. “Knowingly relying” means you have actual knowledge that, but for the patent license, your conveying the covered work in a country, or your recipient's use of the covered work in a country, would infringe one or more identifiable patents in that country that you have reason to believe are valid.

    If, pursuant to or in connection with a single transaction or arrangement, you convey, or propagate by procuring conveyance of, a covered work, and grant a patent license to some of the parties receiving the covered work authorizing them to use, propagate, modify or convey a specific copy of the covered work, then the patent license you grant is automatically extended to all recipients of the covered work and works based on it.

    A patent license is “discriminatory” if it does not include within the scope of its coverage, prohibits the exercise of, or is conditioned on the non-exercise of one or more of the rights that are specifically granted under this License. You may not convey a covered work if you are a party to an arrangement with a third party that is in the business of distributing software, under which you make payment to the third party based on the extent of your activity of conveying the work, and under which the third party grants, to any of the parties who would receive the covered work from you, a discriminatory patent license (a) in connection with copies of the covered work conveyed by you (or copies made from those copies), or (b) primarily for and in connection with specific products or compilations that contain the covered work, unless you entered into that arrangement, or that patent license was granted, prior to 28 March 2007.

    Nothing in this License shall be construed as excluding or limiting any implied license or other defenses to infringement that may otherwise be available to you under applicable patent law.

12. No Surrender of Others' Freedom.

    If conditions are imposed on you (whether by court order, agreement or otherwise) that contradict the conditions of this License, they do not excuse you from the conditions of this License. If you cannot convey a covered work so as to satisfy simultaneously your obligations under this License and any other pertinent obligations, then as a consequence you may not convey it at all. For example, if you agree to terms that obligate you to collect a royalty for further conveying from those to whom you convey the Program, the only way you could satisfy both those terms and this License would be to refrain entirely from conveying the Program.

13. Use with the GNU Affero General Public License.

    Notwithstanding any other provision of this License, you have permission to link or combine any covered work with a work licensed under version 3 of the GNU Affero General Public License into a single combined work, and to convey the resulting work. The terms of this License will continue to apply to the part which is the covered work, but the special requirements of the GNU Affero General Public License, section 13, concerning interaction through a network will apply to the combination as such.

14. Revised Versions of this License.

    The Free Software Foundation may publish revised and/or new versions of the GNU General Public License from time to time. Such new versions will be similar in spirit to the present version, but may differ in detail to address new problems or concerns.

    Each version is given a distinguishing version number. If the Program specifies that a certain numbered version of the GNU General Public License “or any later version” applies to it, you have the option of following the terms and conditions either of that numbered version or of any later version published by the Free Software Foundation. If the Program does not specify a version number of the GNU General Public License, you may choose any version ever published by the Free Software Foundation.

    If the Program specifies that a proxy can decide which future versions of the GNU General Public License can be used, that proxy's public statement of acceptance of a version permanently authorizes you to choose that version for the Program.

    Later license versions may give you additional or different permissions. However, no additional obligations are imposed on any author or copyright holder as a result of your choosing to follow a later version.

15. Disclaimer of Warranty.

    THERE IS NO WARRANTY FOR THE PROGRAM, TO THE EXTENT PERMITTED BY APPLICABLE LAW. EXCEPT WHEN OTHERWISE STATED IN WRITING THE COPYRIGHT HOLDERS AND/OR OTHER PARTIES PROVIDE THE PROGRAM “AS IS” WITHOUT WARRANTY OF ANY KIND, EITHER EXPRESSED OR IMPLIED, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE. THE ENTIRE RISK AS TO THE QUALITY AND PERFORMANCE OF THE PROGRAM IS WITH YOU. SHOULD THE PROGRAM PROVE DEFECTIVE, YOU ASSUME THE COST OF ALL NECESSARY SERVICING, REPAIR OR CORRECTION.

16. Limitation of Liability.

    IN NO EVENT UNLESS REQUIRED BY APPLICABLE LAW OR AGREED TO IN WRITING WILL ANY COPYRIGHT HOLDER, OR ANY OTHER PARTY WHO MODIFIES AND/OR CONVEYS THE PROGRAM AS PERMITTED ABOVE, BE LIABLE TO YOU FOR DAMAGES, INCLUDING ANY GENERAL, SPECIAL, INCIDENTAL OR CONSEQUENTIAL DAMAGES ARISING OUT OF THE USE OR INABILITY TO USE THE PROGRAM (INCLUDING BUT NOT LIMITED TO LOSS OF DATA OR DATA BEING RENDERED INACCURATE OR LOSSES SUSTAINED BY YOU OR THIRD PARTIES OR A FAILURE OF THE PROGRAM TO OPERATE WITH ANY OTHER PROGRAMS), EVEN IF SUCH HOLDER OR OTHER PARTY HAS BEEN ADVISED OF THE POSSIBILITY OF SUCH DAMAGES.

17. Interpretation of Sections 15 and 16.

    If the disclaimer of warranty and limitation of liability provided above cannot be given local legal effect according to their terms, reviewing courts shall apply local law that most closely approximates an absolute waiver of all civil liability in connection with the Program, unless a warranty or assumption of liability accompanies a copy of the Program in return for a fee.

END OF TERMS AND CONDITIONS

How to Apply These Terms to Your New Programs

If you develop a new program, and you want it to be of the greatest possible use to the public, the best way to achieve this is to make it free software which everyone can redistribute and change under these terms.

To do so, attach the following notices to the program. It is safest to attach them to the start of each source file to most effectively state the exclusion of warranty; and each file should have at least the “copyright” line and a pointer to where the full notice is found.

     one line to give the program's name and a brief idea of what it does.
     Copyright (C) year name of author
     
     This program is free software: you can redistribute it and/or modify
     it under the terms of the GNU General Public License as published by
     the Free Software Foundation, either version 3 of the License, or (at
     your option) any later version.
     
     This program is distributed in the hope that it will be useful, but
     WITHOUT ANY WARRANTY; without even the implied warranty of
     MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
     General Public License for more details.
     
     You should have received a copy of the GNU General Public License
     along with this program.  If not, see http://www.gnu.org/licenses/.

Also add information on how to contact you by electronic and paper mail.

If the program does terminal interaction, make it output a short notice like this when it starts in an interactive mode:

     program Copyright (C) year name of author
     This program comes with ABSOLUTELY NO WARRANTY; for details type ‘show w’.
     This is free software, and you are welcome to redistribute it
     under certain conditions; type ‘show c’ for details.

The hypothetical commands ‘show w’ and ‘show c’ should show the appropriate parts of the General Public License. Of course, your program's commands might be different; for a GUI interface, you would use an “about box”.

You should also get your employer (if you work as a programmer) or school, if any, to sign a “copyright disclaimer” for the program, if necessary. For more information on this, and how to apply and follow the GNU GPL, see http://www.gnu.org/licenses/.

The GNU General Public License does not permit incorporating your program into proprietary programs. If your program is a subroutine library, you may consider it more useful to permit linking proprietary applications with the library. If this is what you want to do, use the GNU Lesser General Public License instead of this License. But first, please read http://www.gnu.org/philosophy/why-not-lgpl.html.

Concept Index

• absorption probabilities, DTMC: Mean time to absorption (DTMC)
• approximate MVA: Single Class Models
• asymmetric M/M/m system: The Asymmetric M/M/m System
• asymptotic bounds: Bounds Analysis
• balanced system bounds: Bounds Analysis
• BCMP network: Single Class Models
• birth-death process, CTMC: Birth-death process (CTMC)
• birth-death process, DTMC: Birth-death process (DTMC)
• blocking queueing network: Single Class Models
• bounds, asymptotic: Bounds Analysis
• bounds, balanced system: Bounds Analysis
• bounds, composite: Bounds Analysis
• bounds, geometric: Bounds Analysis
• bounds, PB: Bounds Analysis
• closed multiclass network: Bounds Analysis
• closed network: Bounds Analysis
• closed network: Single Class Models
• closed network, approximate analysis: Single Class Models
• closed network, finite capacity: Single Class Models
• closed network, multiple classes: Generic Algorithms
• closed network, multiple classes: Multiple Class Models
• closed network, multiple classes: Single Class Models
• closed network, single class: Bounds Analysis
• closed network, single class: Generic Algorithms
• closed network, single class: Single Class Models
• CMVA: Single Class Models
• composite bounds: Bounds Analysis
• conditional MVA (CMVA): Single Class Models
• continuous time Markov chain: First passage times (CTMC)
• continuous time Markov chain: Mean time to absorption (CTMC)
• continuous time Markov chain: Time-averaged expected sojourn times (CTMC)
• continuous time Markov chain: Birth-death process (CTMC)
• continuous time Markov chain: State occupancy probabilities (CTMC)
• convolution algorithm: Single Class Models
• copyright: Copying
• CTMC: First passage times (CTMC)
• CTMC: Mean time to absorption (CTMC)
• CTMC: Time-averaged expected sojourn times (CTMC)
• CTMC: Birth-death process (CTMC)
• CTMC: State occupancy probabilities (CTMC)
• deprecated functions: Naming Conventions
• discrete time Markov chain: First passage times (DTMC)
• discrete time Markov chain: Mean time to absorption (DTMC)
• discrete time Markov chain: Time-averaged expected sojourn times (DTMC)
• discrete time Markov chain: Expected number of visits (DTMC)
• discrete time Markov chain: Birth-death process (DTMC)
• discrete time Markov chain: State occupancy probabilities (DTMC)
• discrete time Markov chain: Discrete-Time Markov Chains
• DTMC: First passage times (DTMC)
• DTMC: Mean time to absorption (DTMC)
• DTMC: Time-averaged expected sojourn times (DTMC)
• DTMC: Expected number of visits (DTMC)
• DTMC: Birth-death process (DTMC)
• DTMC: State occupancy probabilities (DTMC)
• DTMC: Discrete-Time Markov Chains
• expected sojourn time, CTMC: Expected sojourn times (CTMC)
• expected sojourn times, DTMC: Expected number of visits (DTMC)
• first passage times: First passage times (DTMC)
• first passage times, CTMC: First passage times (CTMC)
• fundamental matrix: Mean time to absorption (DTMC)
• geometric bounds: Bounds Analysis
• load-dependent service center: Single Class Models
• M/G/1 system: The M/G/1 System
• M/H_m/1 system: The M/Hm/1 System
• M/M/1 system: The M/M/1 System
• M/M/1/K system: The M/M/1/K System
• M/M/inf system: The M/M/inf System
• M/M/m system: The M/M/m System
• M/M/m/K system: The M/M/m/K System
• Markov chain, continuous time: First passage times (CTMC)
• Markov chain, continuous time: Mean time to absorption (CTMC)
• Markov chain, continuous time: Time-averaged expected sojourn times (CTMC)
• Markov chain, continuous time: Expected sojourn times (CTMC)
• Markov chain, continuous time: Birth-death process (CTMC)
• Markov chain, continuous time: State occupancy probabilities (CTMC)
• Markov chain, continuous time: Continuous-Time Markov Chains
• Markov chain, discrete time: First passage times (DTMC)
• Markov chain, discrete time: Mean time to absorption (DTMC)
• Markov chain, discrete time: Time-averaged expected sojourn times (DTMC)
• Markov chain, discrete time: Expected number of visits (DTMC)
• Markov chain, discrete time: Birth-death process (DTMC)
• Markov chain, discrete time: State occupancy probabilities (DTMC)
• Markov chain, discrete time: Discrete-Time Markov Chains
• Markov chain, state occupancy probabilities: State occupancy probabilities (CTMC)
• Markov chain, stationary probabilities: State occupancy probabilities (DTMC)
• Markov chain, transient probabilities: State occupancy probabilities (DTMC)
• mean recurrence times: First passage times (DTMC)
• mean time to absorption, CTMC: Mean time to absorption (CTMC)
• mean time to absorption, DTMC: Mean time to absorption (DTMC)
• Mean Value Analysis, conditional (CMVA): Single Class Models
• Mean Value Analysys (MVA): Multiple Class Models
• Mean Value Analysys (MVA): Single Class Models
• Mean Value Analysys (MVA), approximate: Multiple Class Models
• Mean Value Analysys (MVA), approximate: Single Class Models
• mixed network: Multiple Class Models
• multiclass network, closed: Bounds Analysis
• multiclass network, closed: Multiple Class Models
• multiclass network, open: Bounds Analysis
• multiclass network, open: Multiple Class Models
• MVA: Single Class Models
• MVA, approximate: Multiple Class Models
• MVA, approximate: Single Class Models
• MVABLO: Single Class Models
• normalization constant: Single Class Models
• open network: Bounds Analysis
• open network: Generic Algorithms
• open network, multiple classes: Multiple Class Models
• open network, single class: Single Class Models
• PB bounds: Bounds Analysis
• population mix: Multiple Class Models
• queueing network with blocking: Single Class Models
• queueing networks: Queueing Networks
• RS blocking: Single Class Models
• stationary probabilities: State occupancy probabilities (CTMC)
• time-alveraged sojourn time, CTMC: Time-averaged expected sojourn times (CTMC)
• time-alveraged sojourn time, DTMC: Time-averaged expected sojourn times (DTMC)
• traffic intensity: The M/M/inf System
• warranty: Copying

Function Index

• ctmc: State occupancy probabilities (CTMC)
• ctmcbd: Birth-death process (CTMC)
• ctmcchkQ: Continuous-Time Markov Chains
• ctmcexps: Expected sojourn times (CTMC)
• ctmcfpt: First passage times (CTMC)
• ctmcmtta: Mean time to absorption (CTMC)
• ctmctaexps: Time-averaged expected sojourn times (CTMC)
• dtmc: State occupancy probabilities (DTMC)
• dtmcbd: Birth-death process (DTMC)
• dtmcchkP: Discrete-Time Markov Chains
• dtmcexps: Expected number of visits (DTMC)
• dtmcfpt: First passage times (DTMC)
• dtmcmtta: Mean time to absorption (DTMC)
• dtmctaexps: Time-averaged expected sojourn times (DTMC)
• qnclosed: Generic Algorithms
• qncmaba: Bounds Analysis
• qncmbsb: Bounds Analysis
• qncmcb: Bounds Analysis
• qncmmva: Multiple Class Models
• qncmmvaap: Multiple Class Models
• qncmnpop: Multiple Class Models
• qncmpopmix: Multiple Class Models
• qncmvisits: Multiple Class Models
• qncsaba: Bounds Analysis
• qncsbsb: Bounds Analysis
• qncscmva: Single Class Models
• qncsconv: Single Class Models
• qncsconvld: Single Class Models
• qncsgb: Bounds Analysis
• qncsmva: Single Class Models
• qncsmvaap: Single Class Models
• qncsmvablo: Single Class Models
• qncsmvald: Single Class Models
• qncspb: Bounds Analysis
• qncsvisits: Single Class Models
• qnmarkov: Single Class Models
• qnmix: Multiple Class Models
• qnmknode: Generic Algorithms
• qnom: Multiple Class Models
• qnomaba: Bounds Analysis
• qnomvisits: Multiple Class Models
• qnopen: Generic Algorithms
• qnos: Single Class Models
• qnosaba: Bounds Analysis
• qnosbsb: Bounds Analysis
• qnosvisits: Single Class Models
• qnsolve: Generic Algorithms
• qsammm: The Asymmetric M/M/m System
• qsmg1: The M/G/1 System
• qsmh1: The M/Hm/1 System
• qsmm1: The M/M/1 System
• qsmm1k: The M/M/1/K System
• qsmminf: The M/M/inf System
• qsmmm: The M/M/m System
• qsmmmk: The M/M/m/K System

Author Index

• Akyildiz, I. F.: Single Class Models
• Bard, Y.: Multiple Class Models
• Bolch, G.: Multiple Class Models
• Bolch, G.: Single Class Models
• Bolch, G.: The Asymmetric M/M/m System
• Bolch, G.: The M/M/m/K System
• Bolch, G.: The M/M/inf System
• Bolch, G.: The M/M/m System
• Bolch, G.: The M/M/1 System
• Bolch, G.: Mean time to absorption (CTMC)
• Buzen, J. P.: Single Class Models
• Casale, G.: Bounds Analysis
• Casale, G.: Single Class Models
• de Meer, H.: Multiple Class Models
• de Meer, H.: Single Class Models
• de Meer, H.: The Asymmetric M/M/m System
• de Meer, H.: The M/M/m/K System
• de Meer, H.: The M/M/inf System
• de Meer, H.: The M/M/m System
• de Meer, H.: The M/M/1 System
• de Meer, H.: Mean time to absorption (CTMC)
• Graham, G. S.: Bounds Analysis
• Graham, G. S.: Multiple Class Models
• Graham, G. S.: Single Class Models
• Greiner, S.: Multiple Class Models
• Greiner, S.: Single Class Models
• Greiner, S.: The Asymmetric M/M/m System
• Greiner, S.: The M/M/m/K System
• Greiner, S.: The M/M/inf System
• Greiner, S.: The M/M/m System
• Greiner, S.: The M/M/1 System
• Greiner, S.: Mean time to absorption (CTMC)
• Hsieh, C. H: Bounds Analysis
• Jain, R.: Single Class Models
• Kerola, T.: Bounds Analysis
• Kobayashi, H.: Single Class Models
• Lam, S.: Bounds Analysis
• Lavenberg, S. S.: Multiple Class Models
• Lavenberg, S. S.: Single Class Models
• Lazowska, E. D.: Bounds Analysis
• Lazowska, E. D.: Multiple Class Models
• Lazowska, E. D.: Single Class Models
• Muntz, R. R.: Bounds Analysis
• Reiser, M.: Multiple Class Models
• Reiser, M.: Single Class Models
• Santini, S.: Multiple Class Models
• Schweitzer, P.: Multiple Class Models
• Schwetman, H.: Multiple Class Models
• Schwetman, H.: Single Class Models
• Serazzi, G.: Bounds Analysis
• Sevcik, K. C.: Bounds Analysis
• Sevcik, K. C.: Multiple Class Models
• Sevcik, K. C.: Single Class Models
• Trivedi, K.: Multiple Class Models
• Trivedi, K.: Single Class Models
• Trivedi, K.: The Asymmetric M/M/m System
• Trivedi, K.: The M/M/m/K System
• Trivedi, K.: The M/M/inf System
• Trivedi, K.: The M/M/m System
• Trivedi, K.: The M/M/1 System
• Trivedi, K.: Mean time to absorption (CTMC)
• Wong, E.: Multiple Class Models
• Zahorjan, J.: Bounds Analysis
• Zahorjan, J.: Multiple Class Models
• Zahorjan, J.: Single Class Models