提高PHP MYSQL查询性能还是服务器问题？

我有一个疑问：

SELECT
a.codigo AS code, a.nome AS name, a.especialidade AS expertise, a.telemovel1 AS mobile, a.email AS email, a.codigopostal AS zipcode
FROM table_name_a a
WHERE a.nome != ''
AND a.codigo!='5555555'
AND a.codigo!='705416'


UNION ALL

SELECT
b.cod AS code, b.nome AS name, b.especialidade AS expertise, b.telem1 AS mobile, b.mail AS email, b.cp AS zipcode
FROM table_name_b b
WHERE b.nome != ''
AND b.cod!='5555555'
AND b.cod!='705416'
AND b.cod NOT IN(SELECT assoc.codigo from table_name_a assoc) 

ORDER BY code ASC

表A:+/-6000行（并在增长）

表B:+/-2000行（并在增长）

问题：

SELECT
a.codigo AS code, a.nome AS name, a.especialidade AS expertise, a.telemovel1 AS mobile, a.email AS email, a.codigopostal AS zipcode
FROM table_name_a a
WHERE a.nome != ''
AND a.codigo!='5555555'
AND a.codigo!='705416'


UNION ALL

SELECT
b.cod AS code, b.nome AS name, b.especialidade AS expertise, b.telem1 AS mobile, b.mail AS email, b.cp AS zipcode
FROM table_name_b b
WHERE b.nome != ''
AND b.cod!='5555555'
AND b.cod!='705416'
AND b.cod NOT IN(SELECT assoc.codigo from table_name_a assoc) 

ORDER BY code ASC

当一个或两个用户正在使用服务器对数据库执行多个查询时，此查询运行良好。。。但是当服务器被使用时，比如说在白天，有10个用户执行多个查询，在任何给定的时间大约有+/-1000个客户端执行一些查询，查询来自：

显示第0-29行（总计7138行，查询耗时0.0734秒）

到

显示第0-29行（总计7138行，查询耗时100.0933秒）

提问：

SELECT
a.codigo AS code, a.nome AS name, a.especialidade AS expertise, a.telemovel1 AS mobile, a.email AS email, a.codigopostal AS zipcode
FROM table_name_a a
WHERE a.nome != ''
AND a.codigo!='5555555'
AND a.codigo!='705416'


UNION ALL

SELECT
b.cod AS code, b.nome AS name, b.especialidade AS expertise, b.telem1 AS mobile, b.mail AS email, b.cp AS zipcode
FROM table_name_b b
WHERE b.nome != ''
AND b.cod!='5555555'
AND b.cod!='705416'
AND b.cod NOT IN(SELECT assoc.codigo from table_name_a assoc) 

ORDER BY code ASC

这是查询性能问题还是服务器配置/性能问题

编辑：

SELECT
a.codigo AS code, a.nome AS name, a.especialidade AS expertise, a.telemovel1 AS mobile, a.email AS email, a.codigopostal AS zipcode
FROM table_name_a a
WHERE a.nome != ''
AND a.codigo!='5555555'
AND a.codigo!='705416'


UNION ALL

SELECT
b.cod AS code, b.nome AS name, b.especialidade AS expertise, b.telem1 AS mobile, b.mail AS email, b.cp AS zipcode
FROM table_name_b b
WHERE b.nome != ''
AND b.cod!='5555555'
AND b.cod!='705416'
AND b.cod NOT IN(SELECT assoc.codigo from table_name_a assoc) 

ORDER BY code ASC

---

字段类型：

SELECT
a.codigo AS code, a.nome AS name, a.especialidade AS expertise, a.telemovel1 AS mobile, a.email AS email, a.codigopostal AS zipcode
FROM table_name_a a
WHERE a.nome != ''
AND a.codigo!='5555555'
AND a.codigo!='705416'


UNION ALL

SELECT
b.cod AS code, b.nome AS name, b.especialidade AS expertise, b.telem1 AS mobile, b.mail AS email, b.cp AS zipcode
FROM table_name_b b
WHERE b.nome != ''
AND b.cod!='5555555'
AND b.cod!='705416'
AND b.cod NOT IN(SELECT assoc.codigo from table_name_a assoc) 

ORDER BY code ASC

cod，codigo->int（11）

nome，telemovel1，telem1，特别是Dade，codigopostal，cp->varchar（200）

电子邮件->varchar（400）

解释查询：

SELECT
a.codigo AS code, a.nome AS name, a.especialidade AS expertise, a.telemovel1 AS mobile, a.email AS email, a.codigopostal AS zipcode
FROM table_name_a a
WHERE a.nome != ''
AND a.codigo!='5555555'
AND a.codigo!='705416'


UNION ALL

SELECT
b.cod AS code, b.nome AS name, b.especialidade AS expertise, b.telem1 AS mobile, b.mail AS email, b.cp AS zipcode
FROM table_name_b b
WHERE b.nome != ''
AND b.cod!='5555555'
AND b.cod!='705416'
AND b.cod NOT IN(SELECT assoc.codigo from table_name_a assoc) 

ORDER BY code ASC

---

---

我建议您重写以下内容：

AND b.cod NOT IN (SELECT assoc.codigo from table_name_a assoc) 

为此：

AND NOT EXISTS (SELECT * FROM table_name_a assoc WHERE assoc.codigo = b.cod)

---

重写的查询应该使用更少的内存，为并发进程保留更多的内存。

我建议您重写以下内容：

AND b.cod NOT IN (SELECT assoc.codigo from table_name_a assoc) 

为此：

AND NOT EXISTS (SELECT * FROM table_name_a assoc WHERE assoc.codigo = b.cod)

---

重写的查询应该使用更少的内存，为并发进程保留更多的内存。

要尝试找到如何选择查询，您应该获得查询计划。这将确定使用了哪些索引以及如何改进它。请参阅MySql中的

EXPLAIN

关键字

---

有了这些信息，您可以确保拥有正确的索引。你也可以考虑临时表。

试图找到如何查询查询，你应该得到查询计划。这将确定使用了哪些索引以及如何改进它。请参阅MySql中的

EXPLAIN

关键字

---

有了这些信息，您可以确保拥有正确的索引。也可以考虑临时表。

不测试它，我认为问题是在不在子句中。发布关于查询性能的问题，而不提供列类型和可用索引的详细信息，这表明您不了解代码实际在做什么-因此这可能是查询/架构性能问题。我在NOT IN中也有大量的加载时间，因此建议重写。@dor:我已经更新了问题，以包含您所请求的内容（老实说，我没有时间构建适当的问题，对此表示抱歉！）您是否对查询进行了解释以了解问题所在？如果不进行测试，我想问题出在NOT in子句中。发布了一个关于查询性能的问题，但没有给出列类型和可用索引的详细信息，这表明您不了解代码实际在做什么-因此这可能是查询/架构性能问题。我在NOT IN中也有大量的加载时间，因此建议重写。@dor:我已经更新了问题，以包含您所请求的内容（老实说，我没有时间构建适当的问题，对此表示抱歉！）您是否对查询进行了解释以了解问题所在？@Zuul能否显示您的查询和我的查询的解释输出？@Zuul但您说我的产品比您的原始解决方案慢：）@Zuul能否显示您的查询和我的查询的解释输出？@Zuul但您说我的产品比您的原始解决方案慢:)