fish.payara.monitoring.model

Class PartialDataset

java.lang.Object

fish.payara.monitoring.model.SeriesDataset

fish.payara.monitoring.model.PartialDataset

All Implemented Interfaces:

Serializable

public final class PartialDataset
extends SeriesDataset

Keeps point data for a fixed window size. The implementation is "effectively immutable". While all primitive fields are immutable the sliding window itself is shared. This does not affect the immutability of the observable window as long as the number of added points since the instance was created is less than the window size. An outdated window can be detected though using isOutdated(). In practice the instances can be treated as if they are fully immutable as long as they are monotonically updated and reading is done from the more recent instances. When add(long, long)ing points the window is first filled to it capacity. Since the buffers are twice the window size they start to slide once the size() reaches the capacity(). When the sliding window reaches the end of the buffer the most recent halve is copied to the first half and sliding starts from there again. This keeps copying memory only occur every capacity() points.

Author:

Jan Bernitt

See Also:

Serialized Form

Method Summary

Modifier and Type	Method and Description
SeriesDataset	add(long time, long value)
int	capacity()
int	estimatedBytesMemory()
long	firstTime()
long	getObservedMax()
long	getObservedMin()
BigInteger	getObservedSum()
int	getObservedValueChanges() Note that minimum is 1 (changing from unknown to know value).
int	getStableCount()
long	getStableSince()
boolean	isOutdated()
boolean	isStable()
long	lastTime()
long	lastValue()
long[]	points() Example:
int	size()
long	time(int index)
long	value(int index)

Methods inherited from class fish.payara.monitoring.model.SeriesDataset

equals, getInstance, getObservedAvg, getObservedSince, getObservedValues, getSeries, hashCode, isStableZero, perSecond, toString

Methods inherited from class java.lang.Object

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

Method Detail

isOutdated

public boolean isOutdated()

Specified by:

isOutdated in class SeriesDataset

Returns:

true in case data sharing with updated sets has caused this set to be outdated. This practically does not happen if the sets are used as intended and usually suggests misuse of a programming mistake.

getStableSince

public long getStableSince()

Specified by:

getStableSince in class SeriesDataset

Returns:

the time of the first value that is still same as SeriesDataset.lastValue().

getStableCount

public int getStableCount()

Specified by:

getStableCount in class SeriesDataset

Returns:

the number of recent observed points that all had the same value which is same as SeriesDataset.lastValue()

getObservedMin

public long getObservedMin()

Specified by:

getObservedMin in class SeriesDataset

Returns:

The smallest value observed so far. If no value was observed Long.MAX_VALUE.

getObservedMax

public long getObservedMax()

Specified by:

getObservedMax in class SeriesDataset

Returns:

The largest value observed so far. If no value was observed Long.MIN_VALUE.

getObservedSum

public BigInteger getObservedSum()

Specified by:

getObservedSum in class SeriesDataset

Returns:

sum of all observed values.

getObservedValueChanges

public int getObservedValueChanges()

Description copied from class: SeriesDataset

Note that minimum is 1 (changing from unknown to know value). Zero means no values have been observed yet. Note also that change count of 1 does not imply that the value never did change just that such a change was never observed.

Specified by:

getObservedValueChanges in class SeriesDataset

Returns:

Number of times the value as altered since it has been monitored.

size

public int size()

Specified by:

size in class SeriesDataset

Returns:

the number of actual points in the dataset. This is not the number of observed points but the number of points still known.

capacity

public int capacity()

Specified by:

capacity in class SeriesDataset

Returns:

the maximum number of points in a dataset before adding a new point does remove the oldest point

value

public long value(int index)

time

public long time(int index)

lastValue

public long lastValue()

Specified by:

lastValue in class SeriesDataset

Returns:

the value of the last of the SeriesDataset.points()

firstTime

public long firstTime()

Specified by:

firstTime in class SeriesDataset

Returns:

the time value of the first of the SeriesDataset.points()

lastTime

public long lastTime()

Specified by:

lastTime in class SeriesDataset

Returns:

the time value of the last of the SeriesDataset.points()

points

public long[] points()

Description copied from class: SeriesDataset

Example:

 [t0, v0, t1, v1, t2, v2]
 

Specified by:

points in class SeriesDataset

Returns:

this dataset as flat array with alternating time and value data.

add

public SeriesDataset add(long time,
                         long value)

Specified by:

add in class SeriesDataset

estimatedBytesMemory

public int estimatedBytesMemory()

Specified by:

estimatedBytesMemory in class SeriesDataset

Returns:

the estimated memory in bytes used by this dataset. Since the object layout in memory is a JVM internal this is only a rough estimation based on the fields. References are assumed to use 8 bytes. Padding is not included.

isStable

public boolean isStable()

Overrides:

isStable in class SeriesDataset