Comment on page

APPROX_COUNT_DISTINCT

Approximates the number of distinct non-null input values. 
This function is used as an approximation of COUNT(DISTINCT ...) in order to improve performance. It should be used when there are relatively few (under 1 million) rows, and where the total number of distinct values is high.
Syntax
APPROX_COUNT_DISTINCT(X)
Arguments
X
Type: any type
The values to be counted.
Returns
Type: bigint
Returns the approximate number of distinct non-null input values. 
If all input values are null, zero is returned.
Examples
Transformation job example
SQL
CREATE JOB function_operator_example
    ADD_MISSING_COLUMNS = true	
    AS INSERT INTO default_glue_catalog.upsolver_samples.orders_transformed_data MAP_COLUMNS_BY_NAME
    SELECT 
        LOWER(ordertype) as ordertype,
        ROUND(AVG(nettotal), 2) as avgtotal
    FROM default_glue_catalog.upsolver_samples.orders_raw_data 
    WHERE $commit_time BETWEEN run_start_time() - PARSE_DURATION('1d') AND run_end_time()
    GROUP BY 1
    LIMIT 2;
Query result
ordertype
avgtotal
pickup
1062.14
shipping
1061.91

ordertype	avgtotal
pickup	1062.14
shipping	1061.91

Aggregate functions

APPROX_COUNT_DISTINCT_EACH

Last modified 1yr ago