Question

这是我的查询，但需要很长时间才能完成。

SELECT count(*) AS count_all 
FROM families 
WHERE family_status_cat_id = 1374 
      AND (SELECT COUNT(*) FROM childrens WHERE family_id = families.id) > 1 
      AND unit_id IN (49,33,74,327,329,453,454,469,472,546,547)

有没有办法让它变得更好？

Answer 1

您的查询可能正常使用正确的索引。我建议以下两点：

create index families_fsc_unitid on families(family_status_cat_id, unit_id);
create index children_familyid on children(family_id);

Answer 2

HAVING子句允许您过滤掉不符合某些聚合条件的行

select
    count(*)
from families
inner join
    (
        SELECT 
            family_id 
        FROM childrens as c
        group by family_id 
        having COUNT(*) > 1 -- more than one child in the family
    ) as MultiChild
    on MultiChild.falily_id = families.id

当然，您需要在我省略的其他条件中加回。

获得记录中有超过1个孩子的家庭的数量

2 个答案: