Question

我有一个类似这样的查询，其中联接〜6000个值

SELECT DISTINCT ON(user_id)
                user_id,
                finished_at as last_deposit_date,
                CASE When currency = 'RUB' Then amount_cents  END as last_deposit_amount_cents
            FROM    payments
            JOIN (VALUES (5),(22),(26)) --~6000 values
            AS v(user_id) USING (user_id)
            WHERE action = 'deposit' 
                AND success = 't'
                AND currency IN ('RUB')
            ORDER BY user_id, finished_at DESC

用于查询具有许多值的查询计划：

Unique  (cost=444606.97..449760.44 rows=19276 width=24) (actual time=6129.403..6418.317 rows=5991 loops=1)
  Buffers: shared hit=2386527, temp read=7807 written=7808
  ->  Sort  (cost=444606.97..447183.71 rows=1030695 width=24) (actual time=6129.401..6295.457 rows=1877039 loops=1)
        Sort Key: payments.user_id, payments.finished_at DESC
        Sort Method: external merge  Disk: 62456kB
        Buffers: shared hit=2386527, temp read=7807 written=7808
        ->  Nested Loop  (cost=0.43..341665.35 rows=1030695 width=24) (actual time=0.612..5085.376 rows=1877039 loops=1)
              Buffers: shared hit=2386521
              ->  Values Scan on "*VALUES*"  (cost=0.00..75.00 rows=6000 width=4) (actual time=0.002..4.507 rows=6000 loops=1)
              ->  Index Scan using index_payments_on_user_id on payments  (cost=0.43..54.78 rows=172 width=28) (actual time=0.010..0.793 rows=313 loops=6000)
                    Index Cond: (user_id = "*VALUES*".column1)
                    Filter: (success AND ((action)::text = 'deposit'::text) AND ((currency)::text = 'RUB'::text))
                    Rows Removed by Filter: 85
                    Buffers: shared hit=2386521
Planning time: 5.886 ms
Execution time: 6429.685 ms

我使用PosgreSQL 10.8.0。有没有机会加快此查询的速度？

我尝试用递归替换DISTINCT：

WITH RECURSIVE t AS (
 (SELECT min(user_id) AS user_id FROM payments)
 UNION ALL
 SELECT (SELECT min(user_id) FROM payments  
 WHERE user_id > t.user_id      
 ) AS user_id FROM
t   
  WHERE t.user_id IS NOT NULL
 )
SELECT payments.* FROM t
JOIN (VALUES (5),(22),(26)) --~6000 VALUES
AS v(user_id) USING (user_id)
, LATERAL (
 SELECT user_id,
        finished_at as last_deposit_date,
        CASE When currency = 'RUB' Then amount_cents  END as last_deposit_amount_cents FROM payments            
        WHERE payments.user_id=t.user_id
            AND action = 'deposit' 
        AND success = 't'
        AND currency IN ('RUB')     
        ORDER BY finished_at DESC LIMIT 1
) AS payments

WHERE t.user_id IS NOT NULL;

但是事实证明，它甚至更慢。

哈希联接（成本= 418.67..21807.22行= 3000宽度= 24）（实际时间= 16.804..10843.174行= 5991循环= 1）哈希值：（t.user_id =“ VALUES ”。column1）缓冲区：共享命中= 6396763 CTE吨 ->递归联合（成本= 0.46..53.73行= 101宽度= 8）（实际时间= 0.142..1942.351行= 237029循环= 1）缓冲区：共享命中= 864281 ->结果（成本= 0.46..0.47行= 1宽度= 8）（实际时间= 0.141..0.142行= 1循环= 1）缓冲区：共享命中= 4 InitPlan 3（返回$ 1） ->限制（费用= 0.43..0.46行= 1宽度= 8）（实际时间= 0.138..0.139行= 1循环= 1）缓冲区：共享命中= 4 ->仅索引扫描使用index_payments_on_user_id上的付款payment_2（成本= 0.43..155102.74行= 4858092宽度= 8）（实际时间= 0.137..0.138行= 1循环= 1）索引条件：（user_id不为空）堆获取：0 缓冲区：共享命中= 4 ->在t t_1上进行工作表扫描（成本= 0.00..5.12行= 10宽度= 8）（实际时间= 0.008..0.008行= 1循环= 237029）过滤器：（user_id不为NULL）筛选器删除的行：0 缓冲区：共享命中= 864277 子计划2 ->结果（成本= 0.48..0.49行= 1宽度= 8）（实际时间= 0.007..0.007行= 1循环= 237028）缓冲区：共享命中= 864277 InitPlan 1（返回$ 3） ->限制（费用= 0.43..0.48行= 1宽度= 8）（实际时间= 0.007..0.007行= 1循环= 237028）缓冲区：共享命中= 864277 ->仅索引扫描使用index_payments_on_user_id上的付款payment_1（成本= 0.43..80786.25行= 1619364宽度= 8）（实际时间= 0.007..0.007行= 1循环= 237028）索引条件：（（user_id不为空）AND（user_id> t_1.user_id））堆访存量：46749 缓冲区：共享命中= 864277 ->嵌套循环（成本= 214.94..21498.23行= 100宽度= 32）（实际时间= 0.475..10794.535行= 167333循环= 1）缓冲区：共享命中= 6396757 ->在t上进行CTE扫描（成本= 0.00..2.02行= 100宽度= 8）（实际时间= 0.145..1998.788行= 237028循环= 1）过滤器：（user_id不为NULL）筛选器删除的行：1 缓冲区：共享命中= 864281 ->限制（cost = 214.94..214.94行= 1宽度= 24）（实际时间= 0.037..0.037行= 1循环= 237028）缓冲区：共享命中= 5532476 ->排序（成本= 214.94..215.37行= 172宽度= 24）（实际时间= 0.036..0.036行= 1循环= 237028）排序关键字：payment.finished_at DESC 排序方式：quicksort内存：25kB 缓冲区：共享命中= 5532476 ->使用index_payments_on_user_id进行支付时的索引扫描（成本= 0.43..214.08行= 172宽度= 24）（实际时间= 0.003..0.034行= 15循环= 237028）索引条件：（user_id = t.user_id）过滤器：（成功AND（（操作）::文本='存款'::文本）AND（（货币）::文本='RUB'::文本））筛选器删除的行：6 缓冲区：共享命中= 5532473 ->哈希（成本= 75.00..75.00行= 6000宽度= 4）（实际时间= 2.255..2.255行= 6000循环= 1）存储桶：8192批次：1内存使用量：275kB ->在“ VALUES ”上扫描值（成本= 0.00..75.00行= 6000宽度= 4）（实际时间= 0.004..1.206行= 6000循环= 1）计划时间：7.029毫秒执行时间：10846.774 ms

Answer 1

对于此查询：

SELECT DISTINCT ON (user_id)
       p.user_id,
       p.finished_at as last_deposit_date,
       (CASE WHEN p.currency = 'RUB' THEN p.amount_cents  END) as last_deposit_amount_cents
FROM payments p JOIN
     (VALUES (5),( 22), (26) --~6000 values
     ) v(user_id)
     USING (user_id)
WHERE p.action = 'deposit' AND
      p.success = 't' ND
      p.currency = 'RUB'
ORDER BY p.user_id, p.finished_at DESC;

我不完全理解CASE表达式，因为WHERE正在过滤掉所有其他值。

也就是说，我希望(action, success, currency, user_id, finished_at desc)上的索引会有所帮助。

如何使用DISTINCT ON和JOIN许多值来优化SQL查询？

1 个答案: