Question

我有一个运行中的PostgreSQL查询，但是要花费大量时间才能执行。我需要帮助对其进行优化。

我有：

尽可能删除内部查询。
从查询中删除了不必要的数据。
创建了一个with查询，该查询从头开始获取所需数据

我需要帮助来优化此查询

    with data as (
      select
      e.id,
      e.name,
      t.barcode,
      tt.variant,
      t.cost_cents::decimal / 100 as ticket_cost,
      t.fee_cents::decimal / 100 as booking_fee
    from
      tickets t
      inner join events e on t.event_id = e.id
      inner join ticket_types tt on t.ticket_type_id = tt.id
    where
      t.status = 2
      and e.source in ('source1', 'source2')
    )
    select
      d.name,
      count(distinct d.barcode) as issued,
      (select count(distinct d2.barcode) from data d2 where d2.id = d.id and d2.variant is null) as sold,
      sum(d.ticket_cost) as ticket_revenue,
      sum(d.booking_fee) as booking_fees
    from
      data d
    group by
      id,
      name

Answer 1

使用EXPLAIN更好地检测慢速零件。它将显示所有零件的成本

Answer 2

您可以通过创建适当的索引来加快联接速度。

此外，删除子查询

(select count(distinct d2.barcode) from data d2 where d2.id = d.id and d2.variant is null)

从SELECT子句中添加到d2表的联接，如下所示：

select
  d.name,
  count(distinct d.barcode) as issued,
  count(distinct d2.barcode) as sold,
  sum(d.ticket_cost) as ticket_revenue,
  sum(d.booking_fee) as booking_fees
from
  data d
  left join data d2 on (d2.id = d.id and d2.variant is null)
group by
  d.id,
  d.name

在PostgreSQL中使用查询进行优化

2 个答案: