class TimeSlices

An object that maintains time slice information for gang schedulers

Public Methods

TimeSlices()

default constructor

~TimeSlices()

Destructor

void setTemplateSlice(map<ID, Computer>* computers)

Sets up template slice, from which all new slices will be copied

bool insert(const Token& token, const Packing packing = FIRSTFIT)

Inserts a process into a time slice

bool remove(const Token& token)

Removes a process from a time slice

set<Process*> slotUnification()

Attempts to unify two slots with disjoint computer allocations

void nextSlice(void)

Advances internal iterator which moves us to the next time slice

void prevSlice(SliceItr& itr)

Moves the provided slice iterator backwards

void setMaxMPL(const SIZE newMPL)

Sets the maximum number of time slices that may be used

void setMemoryAware(const bool value)

Sets whether the slicer should care about memory constraints or not

SIZE getNumSlices(void) const

Gets the number of slices currently in use

SIZE getMaxMPL(void) const

Returns the maximum number of slices that may be used

SIZE getCurrentMPL(void) const

Gets the current number of slices that are being used

SIZE getBiggestMPL(void) const

Gets the largest number of slices that have been used

double getAverageMPL(void) const

Gets the average number of slices that have been used

void checkMPL(void)

To calculate average MPL.

bool isInCurrentSlice(Process* process)

Checks if the given process is executing in the current time slice

list<ID> getCurrentlyRunning(void)

Returns the process IDs of processes running in the current time slice.

Documentation

An object that maintains time slice information for gang schedulers

TimeSlices()

~TimeSlices()

void setTemplateSlice(map<ID, Computer>* computers)

bool insert(const Token& token, const Packing packing = FIRSTFIT)

Depending on availability and the packing scheme, this might involve putting the process in an existing slice or creating a new slice to accomodate it.

bool remove(const Token& token)

A side effect of this event may be the removal of the time slice

set<Process*> slotUnification()

If unification is not possible, it still runs some of the processes in the current time slice if possible - which is alternate scheduling

void nextSlice(void)

void prevSlice(SliceItr& itr)

void setMaxMPL(const SIZE newMPL)

0 signifies an infinite number of slices may be used.

void setMemoryAware(const bool value)

If the slicer cares, then it will not allocate computers to processes that would push them over their allocation thresholds.

SIZE getNumSlices(void) const

This is equivalent to the current MPL

SIZE getMaxMPL(void) const

0 translates to an infinite number

SIZE getCurrentMPL(void) const

SIZE getBiggestMPL(void) const

double getAverageMPL(void) const

void checkMPL(void)

bool isInCurrentSlice(Process* process)

list<ID> getCurrentlyRunning(void)

This class has no child classes.

Author:

John Oleszkiewicz

Alphabetic index HTML hierarchy of classes or Java