jmarkov.basic
Class State

java.lang.Object
  jmarkov.basic.State

All Implemented Interfaces:

java.lang.Comparable<State>, JMarkovElement

Direct Known Subclasses:

GeomRelState, GeomState, PropertiesState, StateC, StateEvent

public abstract class State
extends java.lang.Object
implements java.lang.Comparable<State>, JMarkovElement

The Class State represent a state in a MarkovProcess or MDP. The user of the class should estiblish her own coding convention AND code the compareTo method. If the State can be represented with a vector of integers describing its properties, then it might be easier to implement PropertiesState rather than State.

Version:

1.0a

Author:

German Ria�o. Universidad de los Andes.

See Also:

PropertiesState class

Constructor Summary

State()

Method Summary

abstract int	compareTo(State j) The method compareTo should be implemented in order to establish a total ordering among the States.
abstract void	computeMOPs(MarkovProcess<?,?> model) This method should be implemented in order to compute all the measures of performance MOPs.
java.lang.String	description() Returns a String that describes the State.
boolean	equals(java.lang.Object o) If Object is not State it returns false.
int	getIndex()
double	getMOP(int index) Gets the value of this MOP.
double	getMOP(java.lang.String mopName, MarkovProcess<?,?> model) Gets the value of the MOP with this name, by calling getMOP(int)
abstract boolean	isConsistent() This method is called when a state is added to a set, if assertions are enabled.
abstract java.lang.String	label() Returns a (hopefully short) label that descibes the State.
int	setMOP(int index, double value) Sets the value of this MOP.
int	setMOP(MarkovProcess<?,?> model, java.lang.String mopName, double value) Sets the value of the MOP with this name.
java.lang.String	toString() Returns the label.

Methods inherited from class java.lang.Object

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

Constructor Detail

State

public State()

Method Detail

compareTo

public abstract int compareTo(State j)

The method compareTo should be implemented in order to establish a total ordering among the States.

Specified by:

compareTo in interface java.lang.Comparable<State>

Returns:

A positive integer if this is grater then j, negative if this is less then j and 0 if this == j.

See Also:

Comparable.compareTo(Object)

equals

public final boolean equals(java.lang.Object o)

If Object is not State it returns false. Otherwise equals := (compareTo(o)==0)

Specified by:

equals in interface JMarkovElement

Overrides:

equals in class java.lang.Object

Parameters:

o - The Object to compare to.

Returns:

True if the elements are equal.

See Also:

Object.equals(java.lang.Object)

setMOP

public final int setMOP(int index,
                        double value)

Sets the value of this MOP.

Parameters:

index -

value -

Returns:

the index where it was added.

setMOP

public int setMOP(MarkovProcess<?,?> model,
                  java.lang.String mopName,
                  double value)

Sets the value of the MOP with this name. If no MOP with this name exists a new one is declared.

Parameters:

mopName -

model - The model being solved.

value -

Returns:

the index where it was added.

computeMOPs

public abstract void computeMOPs(MarkovProcess<?,?> model)

This method should be implemented in order to compute all the measures of performance MOPs. Inside it you should issue commands like setMop("Utilization server 1", x, model);. * For large models override this method as empty and rather override getMOP(int). Do NOT mix both approaches!!

Parameters:

model - The model being solved.

See Also:

getMOP(int)

getMOP

public final double getMOP(java.lang.String mopName,
                           MarkovProcess<?,?> model)

Gets the value of the MOP with this name, by calling getMOP(int)

Parameters:

mopName - The name of the MOP.

model - Model being solved.

Returns:

current MOP value

See Also:

getMOP(int)

getMOP

public double getMOP(int index)

Gets the value of this MOP. The value should had been set via the setMOP method. Alternatively, for better performance define the MOP Names when implementing the MarkovProcess class and override this method. To define the names in the constructor call the method setMOPs(String[]). The index is the same as the one used in the array in the aforementioned method.

Parameters:

index -

Returns:

The value of this MOP.

See Also:

MarkovProcess.setMOPs(String[])

isConsistent

public abstract boolean isConsistent()

This method is called when a state is added to a set, if assertions are enabled. You should include code that checks the consistency of the paprameters entered. It is very helpful during depelopment. Once assertions are disabled, this will not reduce the speed of your program.

Returns:

true if the state is consistent.

label

public abstract java.lang.String label()

Returns a (hopefully short) label that descibes the State. It is used by all print methods and in the GUI.

Specified by:

label in interface JMarkovElement

Returns:

A shor String label that identifies the state.

See Also:

JMarkovElement.description()

description

public java.lang.String description()

Returns a String that describes the State. By default it is an empty string, but you should implement it in order to get a meaningful description.

Specified by:

description in interface JMarkovElement

Returns:

A String description of the State

See Also:

JMarkovElement.label()

toString

public final java.lang.String toString()

Returns the label.

Specified by:

toString in interface JMarkovElement

Overrides:

toString in class java.lang.Object

Returns:

A String label.

See Also:

label()

getIndex

public final int getIndex()

Returns:

The index in the State set