Package io.micrometer.core.instrument.distribution

Class DistributionStatisticConfig

java.lang.Object

io.micrometer.core.instrument.distribution.DistributionStatisticConfig

All Implemented Interfaces:

Mergeable<DistributionStatisticConfig>

public class DistributionStatisticConfig
extends java.lang.Object
implements Mergeable<DistributionStatisticConfig>

Configures the distribution statistics that emanate from meters like Timer and DistributionSummary.

These statistics include max, percentiles, percentile histograms, and SLA violations.

Many distribution statistics are decayed to give greater weight to recent samples.

Nested Class Summary

Nested Classes

Modifier and Type	Class	Description
static class	DistributionStatisticConfig.Builder

Field Summary

Fields

Modifier and Type	Field	Description
static DistributionStatisticConfig	DEFAULT
static DistributionStatisticConfig	NONE

Constructor Summary

Constructors

Constructor	Description
DistributionStatisticConfig()

Method Summary

Modifier and Type	Method	Description
static DistributionStatisticConfig.Builder	builder()
java.lang.Integer	getBufferLength()	Statistics like max, percentiles, and histogram counts decay over time to give greater weight to recent samples (exception: histogram counts are cumulative for those systems that expect cumulative histogram buckets).
java.time.Duration	getExpiry()	Statistics like max, percentiles, and histogram counts decay over time to give greater weight to recent samples (exception: histogram counts are cumulative for those systems that expect cumulative histogram buckets).
java.util.NavigableSet<java.lang.Long>	getHistogramBuckets(boolean supportsAggregablePercentiles)
java.lang.Long	getMaximumExpectedValue()	The maximum value that the meter is expected to observe.
java.lang.Long	getMinimumExpectedValue()	The minimum value that the meter is expected to observe.
java.lang.Integer	getPercentilePrecision()	Determines the number of digits of precision to maintain on the dynamic range histogram used to compute percentile approximations.
double[]	getPercentiles()	Produces an additional time series for each requested percentile.
long[]	getSlaBoundaries()	Publish at a minimum a histogram containing your defined SLA boundaries.
java.lang.Boolean	isPercentileHistogram()	Adds histogram buckets used to generate aggregable percentile approximations in monitoring systems that have query facilities to do so (e.g.
boolean	isPublishingHistogram()
boolean	isPublishingPercentiles()
DistributionStatisticConfig	merge(DistributionStatisticConfig parent)	Merges two configurations.

Methods inherited from class java.lang.Object

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

Field Detail

DEFAULT

public static final DistributionStatisticConfig DEFAULT

NONE

public static final DistributionStatisticConfig NONE

Constructor Detail

DistributionStatisticConfig

public DistributionStatisticConfig()

Method Detail

builder

public static DistributionStatisticConfig.Builder builder()

merge

public DistributionStatisticConfig merge(DistributionStatisticConfig parent)

Merges two configurations. Any options that are non-null in this configuration take precedence. Any options that are non-null in the parent are used otherwise.

Specified by:

merge in interface Mergeable<DistributionStatisticConfig>

Parameters:

parent - The configuration to merge with. The parent takes lower precedence than this configuration.

Returns:

A new, merged, immutable configuration.

getHistogramBuckets

public java.util.NavigableSet<java.lang.Long> getHistogramBuckets(boolean supportsAggregablePercentiles)

isPercentileHistogram

@Nullable
public java.lang.Boolean isPercentileHistogram()

Adds histogram buckets used to generate aggregable percentile approximations in monitoring systems that have query facilities to do so (e.g. Prometheus' histogram_quantile, Atlas' :percentiles).

Returns:

This builder.

getPercentiles

@Nullable
public double[] getPercentiles()

Produces an additional time series for each requested percentile. This percentile is computed locally, and so can't be aggregated with percentiles computed across other dimensions (e.g. in a different instance). Use percentileHistogram to publish a histogram that can be used to generate aggregable percentile approximations.

Returns:

Percentiles to compute and publish. The 95th percentile should be expressed as 0.95

getPercentilePrecision

@Nullable
public java.lang.Integer getPercentilePrecision()

Determines the number of digits of precision to maintain on the dynamic range histogram used to compute percentile approximations. The higher the degrees of precision, the more accurate the approximation is at the cost of more memory.

Returns:

The digits of precision to maintain for percentile approximations.

getMinimumExpectedValue

@Nullable
public java.lang.Long getMinimumExpectedValue()

The minimum value that the meter is expected to observe. Sets a lower bound on histogram buckets that are shipped to monitoring systems that support aggregable percentile approximations.

Returns:

The minimum value that this distribution summary is expected to observe.

getMaximumExpectedValue

@Nullable
public java.lang.Long getMaximumExpectedValue()

The maximum value that the meter is expected to observe. Sets an upper bound on histogram buckets that are shipped to monitoring systems that support aggregable percentile approximations.

Returns:

The maximum value that the meter is expected to observe.

getExpiry

@Nullable
public java.time.Duration getExpiry()

Statistics like max, percentiles, and histogram counts decay over time to give greater weight to recent samples (exception: histogram counts are cumulative for those systems that expect cumulative histogram buckets). Samples are accumulated to such statistics in ring buffers which rotate after this expiry, with a buffer length of bufferLength.

Returns:

The amount of time samples are accumulated to a histogram before it is reset and rotated.

getBufferLength

@Nullable
public java.lang.Integer getBufferLength()

Statistics like max, percentiles, and histogram counts decay over time to give greater weight to recent samples (exception: histogram counts are cumulative for those systems that expect cumulative histogram buckets). Samples are accumulated to such statistics in ring buffers which rotate after expiry, with this buffer length.

Returns:

The number of histograms to keep in the ring buffer.

getSlaBoundaries

@Nullable
public long[] getSlaBoundaries()

Publish at a minimum a histogram containing your defined SLA boundaries. When used in conjunction with percentileHistogram, the boundaries defined here are included alongside other buckets used to generate aggregable percentile approximations.

Returns:

The SLA boundaries to include the set of histogram buckets shipped to the monitoring system.

isPublishingPercentiles

public boolean isPublishingPercentiles()

isPublishingHistogram

public boolean isPublishingHistogram()