问题描述
几天前，当我有一个小型数据库时，我已经写了一些查询，一切都很好。 但是，此后数据库的大小不断增长，现在需要大量时间才能运行。 有什么方法可以使查询运行更快？
explain SELECT job.id,job_title,text_job_description,html_job_description,amp_html_job_desc,destination,created_at,org.id,org.name,org.logo
FROM jobs.job as job LEFT OUTER JOIN
jobs.organization as org
ON job.organization_id = org.id
WHERE job.is_expired = 0 and job.is_hidden = 0 and
deleted_at is NULL
order by job.created_at desc
offset 0 limit 20;
QUERY PLAN
---------------------------------------------------------------------------------------------
Limit (cost=10514.03..10514.08 rows=20 width=1456)
-> Sort (cost=10514.03..10594.67 rows=32258 width=1456)
Sort Key: job.created_at DESC
-> Hash Left Join (cost=1.20..9655.65 rows=32258 width=1456)
Hash Cond: (job.organization_id = org.id)
-> Seq Scan on job (cost=0.00..9529.90 rows=32258 width=1384)
Filter: ((deleted_at IS NULL) AND (is_expired = 0) AND (is_hidden = 0))
-> Hash (cost=1.09..1.09 rows=9 width=72)
-> Seq Scan on organization org (cost=0.00..1.09 rows=9 width=72)
解决方法
大概您的查询是：
SELECT . . .
FROM jobs.job j LEFT OUTER JOIN
jobs.organization o
ON j.organization_id = o.id
WHERE j.is_expired = 0 and j.is_hidden = 0 and
j.deleted_at is NULL
------^ not identified in your query
ORDER BY j.created_at desc
OFFSET 0 LIMIT 20;
如果是，则需要在(job(is_expired,is_hidden,deled_at,created_at,organization_id)和organization(id)上建立索引。如果将id定义为主键，则可能不需要后者。