postgresql 中的 like 查询优化

当时数量量比较庞大的时候，做模糊查询效率很慢，为了优化查询效率，尝试如下方法做效率对比

一、对比情况说明： 

• 1、数据量100w条数据
• 2、执行sql

 

explain analyze SELECT
	c_patent,
	c_applyissno,
	d_applyissdate,
	d_applydate,
	c_patenttype_dimn,
	c_newlawstatus,
	c_abstract 
FROM
	public.t_knowl_patent_zlxx_temp 
WHERE
	c_applicant LIKE '%abcd%';

1、未建索时执行计划 

Gather  (cost=1000.00..83803.53 rows=92 width=1278) (actual time=217.264..217.264 rows=0 loops=1)
  Workers Planned: 2
  Workers Launched: 2
  ->  Parallel Seq Scan on t_knowl_patent_zlxx  (cost=0.00..82794.33 rows=38 width=1278) (actual time=212.355..212.355 rows=0 loops=3)
        Filter: ((c_applicant)::text ~~ '%abcd%'::text)
        Rows Removed by Filter: 333333
Planning time: 0.272 ms
Execution time: 228.116 ms

 2、btree索引

CREATE INDEX idx_public_t_knowl_patent_zlxx_applicant ON public.t_knowl_patent_zlxx(c_applicant varchar_pattern_ops);

 执行计划

 

Gather  (cost=1000.00..83803.53 rows=92 width=1278) (actual time=208.253..208.253 rows=0 loops=1)
  Workers Planned: 2
  Workers Launched: 2
  ->  Parallel Seq Scan on t_knowl_patent_zlxx  (cost=0.00..82794.33 rows=38 width=1278) (actual time=203.573..203.573 rows=0 loops=3)
        Filter: ((c_applicant)::text ~~ '%abcd%'::text)
        Rows Removed by Filter: 333333
Planning time: 0.116 ms
Execution time: 218.189 ms

 但是如果将查询sql稍微改动一下，把like查询中的前置%去掉是这样的

 

Index Scan using idx_public_t_knowl_patent_zlxx_applicant on t_knowl_patent_zlxx_temp  (cost=0.55..8.57 rows=92 width=1278) (actual time=0.292..0.292 rows=0 loops=1)
  Index Cond: (((c_applicant)::text ~>=~ 'abcd'::text) AND ((c_applicant)::text ~<~ 'abcd'::text))
  Filter: ((c_applicant)::text ~~ 'abcd%'::text)
Planning time: 0.710 ms
Execution time: 0.378 ms

 3、gin索引

创建索引语句(postgresql要求在9.6版本及以上)

 

create extension pg_trgm;
	
CREATE INDEX idx_public_t_knowl_patent_zlxx_applicant ON public.t_knowl_patent_zlxx USING gin (c_applicant gin_trgm_ops);

 执行计划

 

Bitmap Heap Scan on t_knowl_patent_zlxx  (cost=244.71..600.42 rows=91 width=1268) (actual time=0.649..0.649 rows=0 loops=1)
  Recheck Cond: ((c_applicant)::text ~~ '%abcd%'::text)
  ->  Bitmap Index Scan on idx_public_t_knowl_patent_zlxx_applicant  (cost=0.00..244.69 rows=91 width=0) (actual time=0.647..0.647 rows=0 loops=1)
        Index Cond: ((c_applicant)::text ~~ '%abcd%'::text)
Planning time: 0.673 ms
Execution time: 0.740 ms

 三、结论

btree索引可以让后置% "abcd%"的模糊匹配走索引，gin + gp_trgm可以让前后置% "%abcd%" 走索引。但是gin 索引也有弊端，以下情况可能导致无法命中：

 

• 搜索字段少于3个字符时，不会命中索引，这是gin自身机制导致。
• 当搜索字段过长时，比如email检索，可能也不会命中索引，造成原因暂时未知。