相关疑难解决方法(0)

如果它坚持进行顺序扫描,我如何强制Postgres使用索引？

我有一个看起来像这样的表:

CREATE TABLE tracks (id SERIAL, artists JSON);

INSERT INTO tracks (id, artists) 
  VALUES (1, '[{"name": "blink-182"}]');

INSERT INTO tracks (id, artists) 
  VALUES (2, '[{"name": "The Dirty Heads"}, {"name": "Louis Richards"}]');

还有其他几个与此问题无关的列.将它们存储为JSON是有原因的.

我要做的是查找具有特定艺术家姓名(完全匹配)的曲目.

我正在使用此查询:

SELECT * FROM tracks 
  WHERE 'ARTIST NAME' IN
    (SELECT value->>'name' FROM json_array_elements(artists))

例如

SELECT * FROM tracks
  WHERE 'The Dirty Heads' IN 
    (SELECT value->>'name' FROM json_array_elements(artists))

但是,这会进行全表扫描,并且速度不是很快.我尝试使用函数创建GIN索引names_as_array(artists)并使用'ARTIST NAME' = ANY names_as_array(artists),但是不使用索引并且查询实际上明显更慢.

使用PostgreSQL 8.4.9,我对查询的PostgreSQL性能有一个奇怪的问题.此查询正在选择3D卷中的一组点,使用a LEFT OUTER JOIN添加相关ID列,其中存在相关ID.x范围的微小变化可能导致PostgreSQL选择不同的查询计划,执行时间从0.01秒到50秒.这是有问题的查询:

SELECT treenode.id AS id,
       treenode.parent_id AS parentid,
       (treenode.location).x AS x,
       (treenode.location).y AS y,
       (treenode.location).z AS z,
       treenode.confidence AS confidence,
       treenode.user_id AS user_id,
       treenode.radius AS radius,
       ((treenode.location).z - 50) AS z_diff,
       treenode_class_instance.class_instance_id AS skeleton_id
  FROM treenode LEFT OUTER JOIN
         (treenode_class_instance INNER JOIN
          class_instance ON treenode_class_instance.class_instance_id
                                                  = class_instance.id
                            AND class_instance.class_id = 7828307)
       ON (treenode_class_instance.treenode_id = treenode.id
           AND treenode_class_instance.relation_id = 7828321)
  WHERE treenode.project_id = 4
    AND (treenode.location).x >= 8000
    AND (treenode.location).x <= (8000 + 4736) …

在我的 PostgreSQL 11.11 中，我有一列jsonb保存如下对象：

{
   "dynamicFields":[
      {
         "name":"200",
         "hidden":false,
         "subfields":[
            {
               "name":"a",
               "value":"Subfield a"
            },
            {
               "name":"b",
               "value":"Subfield b"
            }           
         ]
      }
   ]
}

dynamicFields是一个数组，subfields也是一个数组，当我点击这样的选择时遇到性能问题：

select *
from my_table a
cross join lateral jsonb_array_elements(jsonb_column -> 'dynamicFields') df
cross join lateral jsonb_array_elements(df -> 'subfields') sf
where df ->> 'name' = '200' and sf ->> 'name' = 'a'

性能问题主要存在于subfield. 我已经添加了这样的索引：

CREATE INDEX idx_my_index ON my_table USING gin ((marc->'dynamicFields') jsonb_path_ops);

如何为subfields内部添加索引dynamicFields？
上面的查询只是一个示例，我在与数据库中其他表的联接中经常使用它。而且我也认识 …

这是explain.depesz.com上的示例计划:

Limit  (cost=65301.950..65301.950 rows=1 width=219) (actual time=886.074..886.074 rows=0 loops=1)
  ->  Sort  (cost=65258.840..65301.950 rows=17243 width=219) (actual time=879.683..885.211 rows=17589 loops=1)
          Sort Key: juliet.romeo
          Sort Method: external merge  Disk: 4664kB
        ->  Hash Join  (cost=30177.210..62214.980 rows=17243 width=219) (actual time=278.986..852.834 rows=17589 loops=1)
                Hash Cond: (whiskey_quebec.whiskey_five = juliet.quebec)
              ->  Bitmap Heap Scan on whiskey_quebec  (cost=326.060..21967.630 rows=17243 width=4) (actual time=7.494..65.956 rows=17589 loops=1)
                      Recheck Cond: (golf = 297)
                    ->  Bitmap Index Scan on kilo  (cost=0.000..321.750 rows=17243 width=0) (actual time=4.638..4.638 rows=17589 loops=1)
                            Index Cond: (golf = 297) …