uniqHLL12 | RawTree

Calculates the approximate number of different argument values, using the [HyperLogLog](https://en.

Calculates the approximate number of different argument values, using the HyperLogLog algorithm.

:::warning We do not recommend using this function. In most cases, use the uniq or uniqCombined function. :::

Implementation details

This function calculates a hash for all parameters in the aggregate, then uses it in calculations. It uses the HyperLogLog algorithm to approximate the number of different argument values.

2^12 5-bit cells are used. The size of the state is slightly more than 2.5 KB. The result is not very accurate (up to ~10% error) for small data sets (<10K elements). However, the result is fairly accurate for high-cardinality data sets (10K-100M), with a maximum error of ~1.6%. Starting from 100M, the estimation error increases, and the function will return very inaccurate results for data sets with extremely high cardinality (1B+ elements).

Provides a determinate result (it does not depend on the query processing order).

Syntax

uniqHLL12(x[, ...])

Arguments

x — The function takes a variable number of parameters. Tuple(T) or Array(T) or Date or DateTime or String or (U)Int* or Float* or Decimal

Returned value

Returns a UInt64-type number representing the approximate number of different argument values. UInt64

Examples

Basic usage

CREATE TABLE example_hll
(
    id UInt32,
    category String
)
ENGINE = Memory;

INSERT INTO example_hll VALUES
(1, 'A'), (2, 'B'), (3, 'A'), (4, 'C'), (5, 'B'), (6, 'A');

SELECT uniqHLL12(category) AS hll_unique_categories
FROM example_hll;

┌─hll_unique_categories─┐
│                     3 │
└───────────────────────┘

Introduced in version 1.1.