Class ElectionStrategy

java.lang.Object

org.tasktide.mutex.strategy.MutexStrategy

org.tasktide.mutex.strategy.ElectionStrategy

public class ElectionStrategy
extends MutexStrategy

Splits NFS ElectionStrategy orientated locking into phases:

1). Initialization.
2). Wait for leadership.
3). Examine iteration:
- Current position.
- Queue stagnation.
- Leader time-to-live.
4). Release lock.

Constructor Summary

Constructors

Constructor	Description
ElectionStrategy()	Construct ElectionStrategy

Method Summary

Modifier and Type	Method	Description
boolean	apply(Mutex mutex)	Apply Mutex
boolean	cleanUp(Mutex mutex)	Remove election files
LoopDecision	evaluateIteration(ElectionState state, Mutex mutex, int limit)	Evaluate iteration
boolean	evaluateLeaderTtl(Path leader)	Evaluate provided leader TTL
PositionCheckValues	examineQueueStagnation(ElectionState state, Mutex mutex, int limit)	Examines current position in queue has changed.
void	initMutex(Mutex mutex)	Initialize Mutex
boolean	release(Mutex mutex)	Release provided Mutex
void	waitUntilLeader(Mutex mutex)	Wait until leadership is acquired

Methods inherited from class MutexStrategy

getStrategyType, inferLeader, inferPosition, queueSize

Methods inherited from class Object

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

Constructor Details

ElectionStrategy

public ElectionStrategy()

Construct ElectionStrategy

Method Details

apply

public boolean apply(Mutex mutex)

Apply Mutex

Specified by:

apply in class MutexStrategy

Parameters:

mutex -

Returns:

boolean

cleanUp

public boolean cleanUp(Mutex mutex)

Remove election files

Specified by:

cleanUp in class MutexStrategy

Parameters:

mutex -

Returns:

boolean

release

public boolean release(Mutex mutex)

Release provided Mutex

Specified by:

release in class MutexStrategy

Parameters:

mutex -

Returns:

boolean

initMutex

public void initMutex(Mutex mutex)

Initialize Mutex

Parameters:

mutex -

examineQueueStagnation

public PositionCheckValues examineQueueStagnation(ElectionState state,
 Mutex mutex,
 int limit)

Examines current position in queue has changed.

1). PositionCheckValues.PROGRESSED = No action
2). PositionCheckValues.UNCHANGED = Increment streak
3). PositionCheckValues.LIMIT_REACHED = Recast ballot

Parameters:

state -

mutex -

limit -

Returns:

PositionCheckValues

evaluateLeaderTtl

public boolean evaluateLeaderTtl(Path leader)

Evaluate provided leader TTL

Parameters:

leader -

Returns:

boolean

evaluateIteration

public LoopDecision evaluateIteration(ElectionState state,
 Mutex mutex,
 int limit)

Evaluate iteration

Parameters:

state -

mutex -

limit -

Returns:

LoopDecision

waitUntilLeader

public void waitUntilLeader(Mutex mutex)

Wait until leadership is acquired

Parameters:

mutex -