Package com.codahale.metrics

Class ExponentiallyDecayingReservoir

java.lang.Object

com.codahale.metrics.ExponentiallyDecayingReservoir

All Implemented Interfaces:

Reservoir

public class ExponentiallyDecayingReservoir
extends java.lang.Object
implements Reservoir

An exponentially-decaying random reservoir of longs. Uses Cormode et al's forward-decaying priority reservoir sampling method to produce a statistically representative sampling reservoir, exponentially biased towards newer entries.

See Also:

Cormode et al. Forward Decay: A Practical Time Decay Model for Streaming Systems. ICDE '09: Proceedings of the 2009 IEEE International Conference on Data Engineering (2009)

Field Summary

Fields

Modifier and Type	Field	Description
private double	alpha
private Clock	clock
private java.util.concurrent.atomic.AtomicLong	count
private static double	DEFAULT_ALPHA
private static int	DEFAULT_SIZE
private java.util.concurrent.locks.ReentrantReadWriteLock	lock
private java.util.concurrent.atomic.AtomicLong	nextScaleTime
private static long	RESCALE_THRESHOLD
private int	size
private long	startTime
private java.util.concurrent.ConcurrentSkipListMap<java.lang.Double,java.lang.Long>	values

Constructor Summary

Constructors

Constructor	Description
ExponentiallyDecayingReservoir()	Creates a new ExponentiallyDecayingReservoir of 1028 elements, which offers a 99.9% confidence level with a 5% margin of error assuming a normal distribution, and an alpha factor of 0.015, which heavily biases the reservoir to the past 5 minutes of measurements.
ExponentiallyDecayingReservoir(int size, double alpha)	Creates a new ExponentiallyDecayingReservoir.
ExponentiallyDecayingReservoir(int size, double alpha, Clock clock)	Creates a new ExponentiallyDecayingReservoir.

Method Summary

Modifier and Type	Method	Description
private long	currentTimeInSeconds()
Snapshot	getSnapshot()	Returns a snapshot of the reservoir's values.
private void	lockForRegularUsage()
private void	lockForRescale()
private void	rescale(long now, long next)
private void	rescaleIfNeeded()
int	size()	Returns the number of values recorded.
private void	unlockForRegularUsage()
private void	unlockForRescale()
void	update(long value)	Adds a new recorded value to the reservoir.
void	update(long value, long timestamp)	Adds an old value with a fixed timestamp to the reservoir.
private double	weight(long t)

Methods inherited from class java.lang.Object

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

Field Detail

DEFAULT_SIZE

private static final int DEFAULT_SIZE

See Also:

Constant Field Values

DEFAULT_ALPHA

private static final double DEFAULT_ALPHA

See Also:

Constant Field Values

RESCALE_THRESHOLD

private static final long RESCALE_THRESHOLD

values

private final java.util.concurrent.ConcurrentSkipListMap<java.lang.Double,java.lang.Long> values

lock

private final java.util.concurrent.locks.ReentrantReadWriteLock lock

alpha

private final double alpha

size

private final int size

count

private final java.util.concurrent.atomic.AtomicLong count

startTime

private volatile long startTime

nextScaleTime

private final java.util.concurrent.atomic.AtomicLong nextScaleTime

clock

private final Clock clock

Constructor Detail

ExponentiallyDecayingReservoir

public ExponentiallyDecayingReservoir()

Creates a new ExponentiallyDecayingReservoir of 1028 elements, which offers a 99.9% confidence level with a 5% margin of error assuming a normal distribution, and an alpha factor of 0.015, which heavily biases the reservoir to the past 5 minutes of measurements.

ExponentiallyDecayingReservoir

public ExponentiallyDecayingReservoir(int size,
                                      double alpha)

Creates a new ExponentiallyDecayingReservoir.

Parameters:

size - the number of samples to keep in the sampling reservoir

alpha - the exponential decay factor; the higher this is, the more biased the reservoir will be towards newer values

ExponentiallyDecayingReservoir

public ExponentiallyDecayingReservoir(int size,
                                      double alpha,
                                      Clock clock)

Creates a new ExponentiallyDecayingReservoir.

Parameters:

size - the number of samples to keep in the sampling reservoir

alpha - the exponential decay factor; the higher this is, the more biased the reservoir will be towards newer values

Method Detail

size

public int size()

Description copied from interface: Reservoir

Returns the number of values recorded.

Specified by:

size in interface Reservoir

Returns:

the number of values recorded

update

public void update(long value)

Description copied from interface: Reservoir

Adds a new recorded value to the reservoir.

Specified by:

update in interface Reservoir

Parameters:

value - a new recorded value

update

public void update(long value,
                   long timestamp)

Adds an old value with a fixed timestamp to the reservoir.

Parameters:

value - the value to be added

timestamp - the epoch timestamp of value in seconds

rescaleIfNeeded

private void rescaleIfNeeded()

getSnapshot

public Snapshot getSnapshot()

Description copied from interface: Reservoir

Returns a snapshot of the reservoir's values.

Specified by:

getSnapshot in interface Reservoir

Returns:

a snapshot of the reservoir's values

currentTimeInSeconds

private long currentTimeInSeconds()

weight

private double weight(long t)

rescale

private void rescale(long now,
                     long next)

unlockForRescale

private void unlockForRescale()

lockForRescale

private void lockForRescale()

lockForRegularUsage

private void lockForRegularUsage()

unlockForRegularUsage

private void unlockForRegularUsage()