Class TreeConvolutionSolver

java.lang.Object

solvers.queueingnet.QNSolver

solvers.queueingnet.treeAlgorithms.treeConvolution.TreeConvolutionSolver

All Implemented Interfaces:

QNSolverInterface

public class TreeConvolutionSolver
extends QNSolver

Class that implements the Tree Convolution algorithm outlined in the paper "A Tree Convolution Algorithm for the Solution of Queueing Networks" (Lam/Lien 1983). Allows solution of sparse networks which are intractable using the sequential convolution algorithm.

Author:

Ben Homer, 2014.

Field Summary

Fields inherited from class QNSolver

G, memUsage, qnm, totalTimer

Constructor Summary

Constructors

Constructor	Description
TreeConvolutionSolver(QNModel qnm, int numThreads)	Constructor.

Method Summary

Modifier and Type	Method	Description
void	computeNormalisingConstant()	Computes G, the normalisation constant, by first running the preprocessor stage (tree planting and complexity evaluation) and then doing a postorder traversal of the tree to calculate G at the root node.d
void	computePerformanceMeasures()	Computes the throughput and mean queue length performance measures from which all other performance measures can be derived.
void	printWelcome()	Prints a short welcome message that says which solver is used.

Methods inherited from class QNSolver

getNormalisingConstant, printMemUsage, printTimeStatistics

Methods inherited from class Object

clone, equals, finalize, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait

Constructor Details

TreeConvolutionSolver

public TreeConvolutionSolver(QNModel qnm,
 int numThreads)
                      throws InternalErrorException

Constructor.

Parameters:

qnm - The queueing network model to solve.

numThreads - The number of threads to solve the model with.

Throws:

InternalErrorException

Method Details

printWelcome

public void printWelcome()

Prints a short welcome message that says which solver is used.

Specified by:

printWelcome in interface QNSolverInterface

Overrides:

printWelcome in class QNSolver

computeNormalisingConstant

public void computeNormalisingConstant()
                                throws InternalErrorException,
OperationNotSupportedException,
InconsistentLinearSystemException,
BTFMatrixErrorException

Computes G, the normalisation constant, by first running the preprocessor stage (tree planting and complexity evaluation) and then doing a postorder traversal of the tree to calculate G at the root node.d

Specified by:

computeNormalisingConstant in interface QNSolverInterface

Overrides:

computeNormalisingConstant in class QNSolver

Throws:

InternalErrorException - * @throws InternalErrorException An exception is thrown if any internal error is encountered during computations.

OperationNotSupportedException

InconsistentLinearSystemException

BTFMatrixErrorException

computePerformanceMeasures

public void computePerformanceMeasures()
                                throws InternalErrorException

Computes the throughput and mean queue length performance measures from which all other performance measures can be derived.

Specified by:

computePerformanceMeasures in interface QNSolverInterface

Overrides:

computePerformanceMeasures in class QNSolver

Throws:

InternalErrorException - Thrown when any computation fails