BucketHistogramStat
Weighted histogram over fixed, caller-chosen bucket edges.
For edges [e0, e1, ..., eK] (strictly increasing) the stat keeps K + 2 buckets: (-inf, e0], (e(i-1), e(i)] for each consecutive pair, and (eK, +inf). Each bucket includes its upper edge, the le convention of Prometheus-style latency buckets. linear builds evenly spaced edges.
Counts are additive, so a negative weight downdates its bucket and a count can go negative; callers must not read one as a population.
Merge adds an interval of the incoming snapshot to the bucket holding its midpoint (or its finite end, for an open interval). A snapshot over the same edges merges exactly; one from another layout, such as SketchResult.toSparseHistogram, is rebinned approximately.
Use cases: distribution-shape monitoring with caller-chosen bin edges, SLO latency buckets, calibration and probability histograms over a known range. Reach for DDSketchStat for unbounded values with logarithmic resolution.
Memory: O(K); one double cell per bucket.
Update: O(log K); binary search over the edges, single atomic add into the bucket.
Concurrency: Per-bucket atomic adds. Exact under every Concurrency level since each update touches exactly one cell.
Constructors
BucketHistogramStat
Properties
bucketEdges
Bucket edges, in increasing order.
concurrency
Thread-safety mode fixed at construction. See Concurrency for the shared contract and the implementation's KDoc for its cell and lock strategy.
Functions
create
Create an independent accumulator at the configured baseline. A null concurrency preserves this accumulator's mode; a non-null value overrides it. Modality subtypes narrow the return type to preserve the input shape.
merge
Fold a compatible snapshot values into this accumulator at timestampNanos. Implementations document whether the combine is exact or approximate and which state the snapshot omits. The source accumulator is unaffected.
Time-dependent statistics, such as windows, place the snapshot at timestampNanos, which must share the clock the caller uses for updates and reads. Time-independent statistics ignore it.
read
Return the current result evaluated at timestampNanos. Time-independent statistics ignore the timestamp. Reading does not add an observation.
Snapshot consistency depends on the implementation's synchronisation. Locks protect coupled state under Concurrency.Strict and Concurrency.HighWrite; independently updated cells and JVM striped adders can still yield a best-effort snapshot. Concurrency.Relaxed permits drift across coupled cells.
reset
Restore the configured baseline, including any prior seed. Preserves the configuration and concurrency.