approx_count_distinct.md

APPROX_COUNT_DISTINCT

description

Syntax

APPROX_COUNT_DISTINCT(expr)

返回类似于 COUNT(DISTINCT col) 结果的近似值聚合函数。

它比 COUNT 和 DISTINCT 组合的速度更快，并使用固定大小的内存，因此对于高基数的列可以使用更少的内存。

example

MySQL > select approx_count_distinct(query_id) from log_statis group by datetime;
+-----------------------------------+
| approx_count_distinct(`query_id`) |
+-----------------------------------+
| 17721                             |
+-----------------------------------+

keyword

APPROX_COUNT_DISTINCT