Package io.opentelemetry.sdk.metrics

Interface Aggregation

public interface Aggregation
Configures how instrument measurements are combined into metrics.

Aggregation provides a set of built-in aggregations via static methods.

Since:
1.14.0

  • Method Details

    • drop

      static Aggregation drop()
      Drops all measurements and don't export any metric.

    • defaultAggregation

      static Aggregation defaultAggregation()
      Choose the default aggregation for the InstrumentType.

    • sum

      static Aggregation sum()
      Aggregates measurements into a MetricDataType.DOUBLE_SUM or MetricDataType.LONG_SUM.

    • lastValue

      static Aggregation lastValue()
      Records the last seen measurement as a MetricDataType.DOUBLE_GAUGE or MetricDataType.LONG_GAUGE.

    • explicitBucketHistogram

      static Aggregation explicitBucketHistogram()
      Aggregates measurements into an explicit bucket MetricDataType.HISTOGRAM using the default bucket boundaries.

    • explicitBucketHistogram

      static Aggregation explicitBucketHistogram(List<Double> bucketBoundaries)
      Aggregates measurements into an explicit bucket MetricDataType.HISTOGRAM.
      Parameters:
      bucketBoundaries - A list of (inclusive) upper bounds for the histogram. Should be in order from lowest to highest.

    • base2ExponentialBucketHistogram

      static Aggregation base2ExponentialBucketHistogram()
      Aggregates measurements into a base-2 MetricDataType.EXPONENTIAL_HISTOGRAM using the default maxBuckets and maxScale.
      Since:
      1.23.0

    • base2ExponentialBucketHistogram

      static Aggregation base2ExponentialBucketHistogram(int maxBuckets, int maxScale)
      Aggregates measurements into a base-2 MetricDataType.EXPONENTIAL_HISTOGRAM.
      Parameters:
      maxBuckets - the max number of positive buckets and negative buckets (max total buckets is 2 * maxBuckets + 1 zero bucket).
      maxScale - the maximum and initial scale. If measurements can't fit in a particular scale given the maxBuckets, the scale is reduced until the measurements can be accommodated. Setting maxScale may reduce the number of downscales. Additionally, the performance of computing bucket index is improved when scale is <= 0.
      Since:
      1.23.0