文斯雜談凸^-^凸

何かの犠牲なしに何も得ることができない、
何かを得るためには同等の代価が必要になる。
            --「鋼の錬金術士」

浅谈ActiveRecord的N + 1查询问题

ORM框架的性能小坑

在使用ActiveRecord这样的ORM工具时,常会嵌套遍历model。
例如,有两个model,Post、Comment,关系是一对多。

1
2
3
4
5
6
7
class Post < ApplicationRecord
  has_many :comments
end

class Comment < ApplicationRecord
  belongs_to :post
end

总共有4个post。

1
2
3
> Post.count
(0.1ms)  SELECT COUNT(*) FROM "posts"
=> 4

获取每个post的所有comment,我们可以:

1
2
3
4
5
6
> Post.all.map{|post| post.comments}
  Post Load (0.3ms)  SELECT "posts".* FROM "posts"
  Comment Load (0.2ms)  SELECT "comments".* FROM "comments" WHERE "comments"."post_id" = ?  [["post_id", 1]]
  Comment Load (0.4ms)  SELECT "comments".* FROM "comments" WHERE "comments"."post_id" = ?  [["post_id", 2]]
  Comment Load (0.6ms)  SELECT "comments".* FROM "comments" WHERE "comments"."post_id" = ?  [["post_id", 3]]
  Comment Load (0.6ms)  SELECT "comments".* FROM "comments" WHERE "comments"."post_id" = ?  [["post_id", 4]]

可以看到为了得到4条数据,我们执行了5(4 + 1)次的查询,这就是所谓N + 1查询问题。

发现问题

除了凭经验外,有一些gem也可以帮助我们提早发现N + 1查询问题。
例如收费的New Relic,免费的Bullet

解决问题

预加载

简单来说,就是提前加载model关系,让ActiveRecord预先加载所需要的数据。
ActiveRecord提供了以下三个方法预加载。

  • includes
  • preload
  • eager_load

他们的区别可以参考这里这里
以最常用的includes方法为例。

1
2
3
> Post.includes(:comments).map{|post| post.comments}
  Post Load (0.2ms)  SELECT "posts".* FROM "posts"
  Comment Load (0.5ms)  SELECT "comments".* FROM "comments" WHERE "comments"."post_id" IN (1, 2, 3, 4)

得到的结果一样,但执行的查询只有两次。

傻瓜式预加载(Goldiloader)

传统预加载的“问题”

includes方法的确很惊艳,但……

第一,代码不够优雅。
例如,假设我们现在想找的是id在1到3之间的post的comment。
一般的我们的逻辑是,查找id在1到3之间的post,获取各post的comment然后合并。
而预加载后的逻辑是,查找id在1到3之间的post,关联comment,再获取各post的comment然后合并。
总觉得有点冗余。

1
2
3
> Post.where(id: 1..3).includes(:comments).map{|post| post.comments}
  Post Load (0.5ms)  SELECT "posts".* FROM "posts" WHERE ("posts"."id" BETWEEN ? AND ?)  [["id", 1], ["id", 3]]
  Comment Load (0.5ms)  SELECT "comments".* FROM "comments" WHERE "comments"."post_id" IN (1, 2, 3)

第二,不符合DRY。
既然我们都不喜欢N + 1,那就应该从源头上杜绝,而不是每次查询时都要主动includes一次。

Goldiloader

懒癌程序员的救星Goldiloader——几乎完美的解决了以上两个问题。

1
gem 'goldiloader'

bundle install以后,就可以用最直接(傻瓜)的方式点点点……

1
2
3
> Post.where(id: 1..3).map{|post| post.comments}
  Post Load (0.2ms)  SELECT "posts".* FROM "posts" WHERE ("posts"."id" BETWEEN ? AND ?)  [["id", 1], ["id", 3]]
  Comment Load (0.3ms)  SELECT "comments".* FROM "comments" WHERE "comments"."post_id" IN (1, 2, 3)
auto_include

Goldiloader默认自动加载所有关联数据,用auto_include: false可以方便地关闭自动加载。

1
2
3
class Post < ApplicationRecord
  has_many :comments, auto_include: false
end
fully_load

以下的方法比较特殊,如果关系已经加载了,则会直接返回已缓存的值,如果没被加载,则会通过SQL查询。

  • first
  • second
  • third
  • fourth
  • fifth
  • forty_two
  • last
  • size
  • ids_reader
  • empty?
  • exists?

假设现在我们需要获取每个post的最新的comment。
但这不是我们想要的。

1
2
3
4
5
6
> Post.all.sum{|post| [post.id, post.comments.last&.content]}
  Post Load (0.1ms)  SELECT "posts".* FROM "posts"
  Comment Load (0.1ms)  SELECT  "comments".* FROM "comments" WHERE "comments"."post_id" = ? ORDER BY "comments"."id" DESC LIMIT ?  [["post_id", 1], ["LIMIT", 1]]
  Comment Load (0.1ms)  SELECT  "comments".* FROM "comments" WHERE "comments"."post_id" = ? ORDER BY "comments"."id" DESC LIMIT ?  [["post_id", 2], ["LIMIT", 1]]
  Comment Load (0.1ms)  SELECT  "comments".* FROM "comments" WHERE "comments"."post_id" = ? ORDER BY "comments"."id" DESC LIMIT ?  [["post_id", 3], ["LIMIT", 1]]
  Comment Load (0.1ms)  SELECT  "comments".* FROM "comments" WHERE "comments"."post_id" = ? ORDER BY "comments"."id" DESC LIMIT ?  [["post_id", 4], ["LIMIT", 1]]

添加选项full_load: true后,当调用上述方法时,Goldiloader会强制自动加载所需的关系。

1
2
3
class Post < ApplicationRecord
  has_many :comments, fully_load: true
end

这才是我们想要的。

1
2
3
> Post.all.sum{|post| [post.id, post.comments.last&.content]}
  Post Load (0.3ms)  SELECT "posts".* FROM "posts"
  Comment Load (0.2ms)  SELECT "comments".* FROM "comments" WHERE "comments"."post_id" IN (1, 2, 3, 4)

Goldiloader也不是万能的

has_one使用SQL limit时的隐患

Goldiloader是ActiveRecord的衍生工具,所以ActiveRecord预加载的副作用也一并继承了。
现在我们自定义一个has_one关系,用以获取最新的一条comment。

1
2
3
4
class Post < ApplicationRecord
  has_many :comments, fully_load: true
  has_one :latest_comment, -> { order(created_at: :desc) }, class_name: 'Comment'
end

遍历post获取最新的comment。

1
> Post.all.map{|post| post.latest_comment}

不使用Goldiloader或者预加载时,每条SQL自动回加上limit 1

1
2
3
4
5
Post Load (0.3ms)  SELECT "posts".* FROM "posts"
Comment Load (0.2ms)  SELECT  "comments".* FROM "comments" WHERE "comments"."post_id" = ? ORDER BY "comments"."created_at" DESC LIMIT ?  [["post_id", 1], ["LIMIT", 1]]
Comment Load (0.2ms)  SELECT  "comments".* FROM "comments" WHERE "comments"."post_id" = ? ORDER BY "comments"."created_at" DESC LIMIT ?  [["post_id", 2], ["LIMIT", 1]]
Comment Load (0.1ms)  SELECT  "comments".* FROM "comments" WHERE "comments"."post_id" = ? ORDER BY "comments"."created_at" DESC LIMIT ?  [["post_id", 3], ["LIMIT", 1]]
Comment Load (0.1ms)  SELECT  "comments".* FROM "comments" WHERE "comments"."post_id" = ? ORDER BY "comments"."created_at" DESC LIMIT ?  [["post_id", 4], ["LIMIT", 1]]

使用Goldiloader或者预加载时,世界变清净了,但同时会有性能隐患,因为post的数据量可能非常大。

1
2
  Post Load (0.5ms)  SELECT "posts".* FROM "posts"
  Comment Load (0.2ms)  SELECT "comments".* FROM "comments" WHERE "comments"."post_id" IN (1, 2, 3, 4) ORDER BY "comments"."created_at" DESC
其他限制

遇到以下的关系(方法),Goldiloader会自动关闭自动预加载。

  • limit
  • offset
  • finder_sql
  • group (due to a Rails bug)
  • from (due to a Rails bug)
  • joins (only Rails 4.0/4.1 – due to a Rails bug)
  • uniq (only Rails 3.2 – due to a Rails bug)

本文结束之前

N + 1查询问题是一个容易被忽略的问题。
发现解决它也不难,includes已经够用,Goldiloader更是锦上添花,对新手足够友好。
不过对于我这种被Rails“坑”习惯的斯德哥尔摩症候群患者来说,没有includes反而没安全感了>_<|||

参考文档