com.yammer.metrics.core
Class HistogramMetric

java.lang.Object
  com.yammer.metrics.core.HistogramMetric

All Implemented Interfaces:

Metric, Percentiled, Summarized

public class HistogramMetric
extends Object
implements Metric, Percentiled, Summarized

A metric which calculates the distribution of a value.

See Also:

Accurately computing running variance

Nested Class Summary

static class

HistogramMetric.SampleType The type of sampling the histogram should be performing.

Constructor Summary

HistogramMetric(HistogramMetric.SampleType type)
Creates a new HistogramMetric with the given sample type.

HistogramMetric(Sample sample)
Creates a new HistogramMetric with the given sample.

Method Summary

void	clear() Clears all recorded values.
long	count() Returns the number of values recorded.
void	dump(File output) Writes the values of the histogram's sample to the given file.
double	max() Returns the largest recorded value.
double	mean() Returns the arithmetic mean of all recorded values.
double	min() Returns the smallest recorded value.
double	percentile(double percentile) Returns the value at the given percentile.
Double[]	percentiles(Double... percentiles) Returns an array of values at the given percentiles.
<T> void	processWith(MetricsProcessor<T> processor, MetricName name, T context)
double	stdDev() Returns the standard deviation of all recorded values.
void	update(int value) Adds a recorded value.
void	update(long value) Adds a recorded value.
List<Long>	values() Returns a list of all values in the histogram's sample.

Methods inherited from class java.lang.Object

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

Constructor Detail

HistogramMetric

public HistogramMetric(HistogramMetric.SampleType type)

Creates a new HistogramMetric with the given sample type.

Parameters:

type - the type of sample to use

HistogramMetric

public HistogramMetric(Sample sample)

Creates a new HistogramMetric with the given sample.

Parameters:

sample - the sample to create a histogram from

Method Detail

clear

public void clear()

Clears all recorded values.

update

public void update(int value)

Adds a recorded value.

Parameters:

value - the length of the value

update

public void update(long value)

Adds a recorded value.

Parameters:

value - the length of the value

count

public long count()

Returns the number of values recorded.

Returns:

the number of values recorded

max

public double max()

Description copied from interface: Summarized

Returns the largest recorded value.

Specified by:

max in interface Summarized

Returns:

the largest recorded value

min

public double min()

Description copied from interface: Summarized

Returns the smallest recorded value.

Specified by:

min in interface Summarized

Returns:

the smallest recorded value

mean

public double mean()

Description copied from interface: Summarized

Returns the arithmetic mean of all recorded values.

Specified by:

mean in interface Summarized

Returns:

the arithmetic mean of all recorded values

stdDev

public double stdDev()

Description copied from interface: Summarized

Returns the standard deviation of all recorded values.

Specified by:

stdDev in interface Summarized

Returns:

the standard deviation of all recorded values

percentile

public double percentile(double percentile)

Returns the value at the given percentile.

Specified by:

percentile in interface Percentiled

Parameters:

percentile - a percentile (0..1)

Returns:

the value at the given percentile

percentiles

public Double[] percentiles(Double... percentiles)

Returns an array of values at the given percentiles.

Specified by:

percentiles in interface Percentiled

Parameters:

percentiles - one or more percentiles (0..1)

Returns:

an array of values at the given percentiles

values

public List<Long> values()

Returns a list of all values in the histogram's sample.

Returns:

a list of all values in the histogram's sample

dump

public void dump(File output)
          throws IOException

Writes the values of the histogram's sample to the given file.

Parameters:

output - the file to which the values will be written

Throws:

IOException - if there is an error writing the values

processWith

public <T> void processWith(MetricsProcessor<T> processor,
                            MetricName name,
                            T context)
                 throws Exception

Specified by:

processWith in interface Metric

Throws:

Exception