这是一个复杂的问题,所以在我设置场景时请耐心等待:
假设我们有一个简化的表格设置,如下所示:
table 1(employee): {
employee_id, -primary key
first_name,
last_name,
days_of_employment
}
with data:
employee_id first_name last_name days_of_employment
111 Jack Stevens 543
222 Clarice Bobber 323
333 Roy Cook 736
444 Fred Roberts 1000
...
table 2(teams): {
team_code, --primary key
description
}
with data:
team_code description
ERA Enrollment Records Assoc.
RR Rolling Runners
FR French Revolution
...
table 3(employees_teams):{
employee_id, --primary key
team_code --primary key
}
with data:
employee_id team_code
111 RR
111 FR
222 FR
222 ERA
333 FR
...
我希望这些表格应该清楚它们是什么以及它们的目的。这是我的需求场景:“我想要 Rolling Runners and Enrollment Records Assoc. 团队中员工的平均就业天数。” 我知道如何编写此查询有两种方法,它们似乎都运行良好,但我真正想知道的是哪一种方法对 oracle 数据库的处理速度更快。请记住,这些查询的编写方式是为了避免产生重复的行,这会破坏平均计算:
查询一:
SELECT AVG(e.days_of_employment) avg_days_of_employment
FROM employee e,
(
SELECT DISTINCT employee_id
FROM employees_teams
WHERE team_code IN ('ERA','RR')) available_employees
WHERE e.employee_id = available_employees.employee_id
查询 2:
SELECT AVG(e.days_of_employment) avg_days_of_employment
FROM employee e
WHERE EXISTS(
SELECT 1
FROM employees_teams et
WHERE et.team_code IN ('ERA','RR')
AND et.employee_id = e.employee_id)
有可能使用我提供的这个示例数据,这种情况一开始可能没有意义,但我仍然想知道哪个查询“更好”使用。