The "N+1 selects problem" is generally stated as a problem in Object-Relational mapping (ORM) discussions, and I understand that it has something to do with having to make a lot of database queries for something that seems simple in the object world.
Does anybody have a more detailed explanation of the problem?
19 Answers 19
Let's say you have a collection of Car objects (database rows), and each Car has a collection of Wheel objects (also rows). In other words, Car → Wheel is a 1-to-many relationship.
Now, let's say you need to iterate through all the cars, and for each one, print out a list of the wheels. The naive O/R implementation would do the following:
SELECT * FROM Cars;
And then for each Car:
SELECT * FROM Wheel WHERE CarId = ?
In other words, you have one select for the Cars, and then N additional selects, where N is the total number of cars.
Alternatively, one could get all wheels and perform the lookups in memory:
SELECT * FROM Wheel;
This reduces the number of round-trips to the database from N+1 to 2. Most ORM tools give you several ways to prevent N+1 selects.
Reference: Java Persistence with Hibernate , chapter 13.
32 Comments
SELECT * from Wheel;), instead of N+1. With a large N, the performance hit can be very significant.What is the N+1 query problem?
The N+1 query problem happens when the data access framework executed N additional SQL statements to fetch the same data that could have been retrieved when executing the primary SQL query.
The larger the value of N, the more queries will be executed, and the larger the performance impact. And, unlike the slow query log that can help you find slow running queries, the N+1 issue won’t be spotted because each individual additional query runs sufficiently fast to not trigger the slow query log.
The problem is executing a large number of additional queries that, overall, take sufficient time to slow down response time.
Let’s consider we have the following post and post_comments database tables which form a one-to-many table relationship:
The post and post_comments tables
We are going to create the following 4 post rows:
INSERT INTO post (title, id)
VALUES ('High-Performance Java Persistence - Part 1', 1)
INSERT INTO post (title, id)
VALUES ('High-Performance Java Persistence - Part 2', 2)
INSERT INTO post (title, id)
VALUES ('High-Performance Java Persistence - Part 3', 3)
INSERT INTO post (title, id)
VALUES ('High-Performance Java Persistence - Part 4', 4)
And, we will also create 4 post_comment child records:
INSERT INTO post_comment (post_id, review, id)
VALUES (1, 'Excellent book to understand Java Persistence', 1)
INSERT INTO post_comment (post_id, review, id)
VALUES (2, 'Must-read for Java developers', 2)
INSERT INTO post_comment (post_id, review, id)
VALUES (3, 'Five Stars', 3)
INSERT INTO post_comment (post_id, review, id)
VALUES (4, 'A great reference book', 4)
N+1 query problem with plain SQL
If you select the post_comments using this SQL query:
List<Tuple> comments = entityManager.createNativeQuery("""
SELECT
pc.id AS id,
pc.review AS review,
pc.post_id AS postId
FROM post_comment pc
""", Tuple.class)
.getResultList();
And, later, you decide to fetch the associated post title for each post_comment:
for (Tuple comment : comments) {
String review = (String) comment.get("review");
Long postId = ((Number) comment.get("postId")).longValue();
String postTitle = (String) entityManager.createNativeQuery("""
SELECT
p.title
FROM post p
WHERE p.id = :postId
""")
.setParameter("postId", postId)
.getSingleResult();
LOGGER.info(
"The Post '{}' got this review '{}'",
postTitle,
review
);
}
You are going to trigger the N+1 query issue because, instead of one SQL query, you executed 5 (1 + 4):
SELECT
pc.id AS id,
pc.review AS review,
pc.post_id AS postId
FROM post_comment pc
SELECT p.title FROM post p WHERE p.id = 1
-- The Post 'High-Performance Java Persistence - Part 1' got this review
-- 'Excellent book to understand Java Persistence'
SELECT p.title FROM post p WHERE p.id = 2
-- The Post 'High-Performance Java Persistence - Part 2' got this review
-- 'Must-read for Java developers'
SELECT p.title FROM post p WHERE p.id = 3
-- The Post 'High-Performance Java Persistence - Part 3' got this review
-- 'Five Stars'
SELECT p.title FROM post p WHERE p.id = 4
-- The Post 'High-Performance Java Persistence - Part 4' got this review
-- 'A great reference book'
Fixing the N+1 query issue is very easy. All you need to do is extract all the data you need in the original SQL query, like this:
List<Tuple> comments = entityManager.createNativeQuery("""
SELECT
pc.id AS id,
pc.review AS review,
p.title AS postTitle
FROM post_comment pc
JOIN post p ON pc.post_id = p.id
""", Tuple.class)
.getResultList();
for (Tuple comment : comments) {
String review = (String) comment.get("review");
String postTitle = (String) comment.get("postTitle");
LOGGER.info(
"The Post '{}' got this review '{}'",
postTitle,
review
);
}
This time, only one SQL query is executed to fetch all the data we are further interested in using.
N+1 query problem with JPA and Hibernate
When using JPA and Hibernate, there are several ways you can trigger the N+1 query issue, so it’s very important to know how you can avoid these situations.
For the next examples, consider we are mapping the post and post_comments tables to the following entities:
The JPA mappings look like this:
@Entity(name = "Post")
@Table(name = "post")
public class Post {
@Id
private Long id;
private String title;
//Getters and setters omitted for brevity
}
@Entity(name = "PostComment")
@Table(name = "post_comment")
public class PostComment {
@Id
private Long id;
@ManyToOne
private Post post;
private String review;
//Getters and setters omitted for brevity
}
FetchType.EAGER
Using FetchType.EAGER either implicitly or explicitly for your JPA associations is a bad idea because you are going to fetch way more data that you need. More, the FetchType.EAGER strategy is also prone to N+1 query issues.
Unfortunately, the @ManyToOne and @OneToOne associations use FetchType.EAGER by default, so if your mappings look like this:
@ManyToOne
private Post post;
You are using the FetchType.EAGER strategy, and, every time you forget to use JOIN FETCH when loading some PostComment entities with a JPQL or Criteria API query:
List<PostComment> comments = entityManager
.createQuery("""
select pc
from PostComment pc
""", PostComment.class)
.getResultList();
You are going to trigger the N+1 query issue:
SELECT
pc.id AS id1_1_,
pc.post_id AS post_id3_1_,
pc.review AS review2_1_
FROM
post_comment pc
SELECT p.id AS id1_0_0_, p.title AS title2_0_0_ FROM post p WHERE p.id = 1
SELECT p.id AS id1_0_0_, p.title AS title2_0_0_ FROM post p WHERE p.id = 2
SELECT p.id AS id1_0_0_, p.title AS title2_0_0_ FROM post p WHERE p.id = 3
SELECT p.id AS id1_0_0_, p.title AS title2_0_0_ FROM post p WHERE p.id = 4
Notice the additional SELECT statements that are executed because the post association has to be fetched prior to returning the List of PostComment entities.
Unlike the default fetch plan, which you are using when calling the find method of the EntityManager, a JPQL or Criteria API query defines an explicit plan that Hibernate cannot change by injecting a JOIN FETCH automatically. So, you need to do it manually.
If you didn't need the post association at all, you are out of luck when using FetchType.EAGER because there is no way to avoid fetching it. That's why it's better to use FetchType.LAZY by default.
But, if you wanted to use post association, then you can use JOIN FETCH to avoid the N+1 query problem:
List<PostComment> comments = entityManager.createQuery("""
select pc
from PostComment pc
join fetch pc.post p
""", PostComment.class)
.getResultList();
for(PostComment comment : comments) {
LOGGER.info(
"The Post '{}' got this review '{}'",
comment.getPost().getTitle(),
comment.getReview()
);
}
This time, Hibernate will execute a single SQL statement:
SELECT
pc.id as id1_1_0_,
pc.post_id as post_id3_1_0_,
pc.review as review2_1_0_,
p.id as id1_0_1_,
p.title as title2_0_1_
FROM
post_comment pc
INNER JOIN
post p ON pc.post_id = p.id
-- The Post 'High-Performance Java Persistence - Part 1' got this review
-- 'Excellent book to understand Java Persistence'
-- The Post 'High-Performance Java Persistence - Part 2' got this review
-- 'Must-read for Java developers'
-- The Post 'High-Performance Java Persistence - Part 3' got this review
-- 'Five Stars'
-- The Post 'High-Performance Java Persistence - Part 4' got this review
-- 'A great reference book'
FetchType.LAZY
Even if you switch to using FetchType.LAZY explicitly for all associations, you can still bump into the N+1 issue.
This time, the post association is mapped like this:
@ManyToOne(fetch = FetchType.LAZY)
private Post post;
Now, when you fetch the PostComment entities:
List<PostComment> comments = entityManager
.createQuery("""
select pc
from PostComment pc
""", PostComment.class)
.getResultList();
Hibernate will execute a single SQL statement:
SELECT
pc.id AS id1_1_,
pc.post_id AS post_id3_1_,
pc.review AS review2_1_
FROM
post_comment pc
But, if afterward, you are going to reference the lazy-loaded post association:
for(PostComment comment : comments) {
LOGGER.info(
"The Post '{}' got this review '{}'",
comment.getPost().getTitle(),
comment.getReview()
);
}
You will get the N+1 query issue:
SELECT p.id AS id1_0_0_, p.title AS title2_0_0_ FROM post p WHERE p.id = 1
-- The Post 'High-Performance Java Persistence - Part 1' got this review
-- 'Excellent book to understand Java Persistence'
SELECT p.id AS id1_0_0_, p.title AS title2_0_0_ FROM post p WHERE p.id = 2
-- The Post 'High-Performance Java Persistence - Part 2' got this review
-- 'Must-read for Java developers'
SELECT p.id AS id1_0_0_, p.title AS title2_0_0_ FROM post p WHERE p.id = 3
-- The Post 'High-Performance Java Persistence - Part 3' got this review
-- 'Five Stars'
SELECT p.id AS id1_0_0_, p.title AS title2_0_0_ FROM post p WHERE p.id = 4
-- The Post 'High-Performance Java Persistence - Part 4' got this review
-- 'A great reference book'
Because the post association is fetched lazily, a secondary SQL statement will be executed when accessing the lazy association in order to build the log message.
Again, the fix consists in adding a JOIN FETCH clause to the JPQL query:
List<PostComment> comments = entityManager.createQuery("""
select pc
from PostComment pc
join fetch pc.post p
""", PostComment.class)
.getResultList();
for(PostComment comment : comments) {
LOGGER.info(
"The Post '{}' got this review '{}'",
comment.getPost().getTitle(),
comment.getReview()
);
}
And, just like in the FetchType.EAGER example, this JPQL query will generate a single SQL statement.
Even if you are using
FetchType.LAZYand don't reference the child association of a bidirectional@OneToOneJPA relationship, you can still trigger the N+1 query issue.
How to automatically detect the N+1 query issue
If you want to automatically detect N+1 query issue in your data access layer, you can use the db-util open-source project.
First, you need to add the following Maven dependency:
<dependency>
<groupId>com.vladmihalcea</groupId>
<artifactId>db-util</artifactId>
<version>${db-util.version}</version>
</dependency>
Afterward, you just have to use SQLStatementCountValidator utility to assert the underlying SQL statements that get generated:
SQLStatementCountValidator.reset();
List<PostComment> comments = entityManager.createQuery("""
select pc
from PostComment pc
""", PostComment.class)
.getResultList();
SQLStatementCountValidator.assertSelectCount(1);
In case you are using FetchType.EAGER and run the above test case, you will get the following test case failure:
SELECT
pc.id as id1_1_,
pc.post_id as post_id3_1_,
pc.review as review2_1_
FROM
post_comment pc
SELECT p.id as id1_0_0_, p.title as title2_0_0_ FROM post p WHERE p.id = 1
SELECT p.id as id1_0_0_, p.title as title2_0_0_ FROM post p WHERE p.id = 2
-- SQLStatementCountMismatchException: Expected 1 statement(s) but recorded 3 instead!
13 Comments
SELECT cars, wheels FROM cars JOIN wheels LIMIT 0, 5. But what you get is 2 cars with 5 wheels (first car with all 4 wheels and second car with only 1 wheel), because LIMIT will limit the entire resultset, not only root clause.join fetch?SELECT
table1.*
, table2.*
INNER JOIN table2 ON table2.SomeFkId = table1.SomeId
That gets you a result set where child rows in table2 cause duplication by returning the table1 results for each child row in table2. O/R mappers should differentiate table1 instances based on a unique key field, then use all the table2 columns to populate child instances.
SELECT table1.*
SELECT table2.* WHERE SomeFkId = #
The N+1 is where the first query populates the primary object and the second query populates all the child objects for each of the unique primary objects returned.
Consider:
class House
{
int Id { get; set; }
string Address { get; set; }
Person[] Inhabitants { get; set; }
}
class Person
{
string Name { get; set; }
int HouseId { get; set; }
}
and tables with a similar structure. A single query for the address "22 Valley St" may return:
Id Address Name HouseId
1 22 Valley St Dave 1
1 22 Valley St John 1
1 22 Valley St Mike 1
The O/RM should fill an instance of Home with ID=1, Address="22 Valley St" and then populate the Inhabitants array with People instances for Dave, John, and Mike with just one query.
A N+1 query for the same address used above would result in:
Id Address
1 22 Valley St
with a separate query like
SELECT * FROM Person WHERE HouseId = 1
and resulting in a separate data set like
Name HouseId
Dave 1
John 1
Mike 1
and the final result being the same as above with the single query.
The advantages to single select is that you get all the data up front which may be what you ultimately desire. The advantages to N+1 is query complexity is reduced and you can use lazy loading where the child result sets are only loaded upon first request.
22 Comments
Supplier with a one-to-many relationship with Product. One Supplier has (supplies) many Products.
***** Table: Supplier *****
+-----+-------------------+
| ID | NAME |
+-----+-------------------+
| 1 | Supplier Name 1 |
| 2 | Supplier Name 2 |
| 3 | Supplier Name 3 |
| 4 | Supplier Name 4 |
+-----+-------------------+
***** Table: Product *****
+-----+-----------+--------------------+-------+------------+
| ID | NAME | DESCRIPTION | PRICE | SUPPLIERID |
+-----+-----------+--------------------+-------+------------+
|1 | Product 1 | Name for Product 1 | 2.0 | 1 |
|2 | Product 2 | Name for Product 2 | 22.0 | 1 |
|3 | Product 3 | Name for Product 3 | 30.0 | 2 |
|4 | Product 4 | Name for Product 4 | 7.0 | 3 |
+-----+-----------+--------------------+-------+------------+
Factors:
Lazy mode for Supplier set to "true" (default)
Fetch mode used for querying on Product is Select
Fetch mode (default): Supplier information is accessed
Caching does not play a role for the first time the
Supplier is accessed
Fetch mode is Select Fetch (default)
// It takes Select fetch mode as a default
Query query = session.createQuery( "from Product p");
List list = query.list();
// Supplier is being accessed
displayProductsListWithSupplierName(results);
select ... various field names ... from PRODUCT
select ... various field names ... from SUPPLIER where SUPPLIER.id=?
select ... various field names ... from SUPPLIER where SUPPLIER.id=?
select ... various field names ... from SUPPLIER where SUPPLIER.id=?
Result:
- 1 select statement for Product
- N select statements for Supplier
This is N+1 select problem!
3 Comments
A good explanation of the problem can be found in the Phabricator documentation
TL;DR
It is much faster to issue 1 query which returns 100 results than to issue 100 queries which each return 1 result.
Load all your data before iterating through it.
More in-detail
The N+1 query problem is a common performance antipattern. It looks like this:
$cats = load_cats(); foreach ($cats as $cat) { $cats_hats => load_hats_for_cat($cat); // ... }Assuming
load_cats()has an implementation that boils down to:SELECT * FROM cat WHERE .....and
load_hats_for_cat($cat)has an implementation something like this:SELECT * FROM hat WHERE catID = .....you will issue "N+1" queries when the code executes, where N is the number of cats:
SELECT * FROM cat WHERE ... SELECT * FROM hat WHERE catID = 1 SELECT * FROM hat WHERE catID = 2 SELECT * FROM hat WHERE catID = 3 SELECT * FROM hat WHERE catID = 4 ...
1 Comment
I can't comment directly on other answers, because I don't have enough reputation. But it's worth noting that the problem essentially only arises because, historically, a lot of dbms have been quite poor when it comes to handling joins (MySQL being a particularly noteworthy example). So n+1 has, often, been notably faster than a join. And then there are ways to improve on n+1 but still without needing a join, which is what the original problem relates to.
However, MySQL is now a lot better than it used to be when it comes to joins. When I first learned MySQL, I used joins a lot. Then I discovered how slow they are, and switched to n+1 in the code instead. But, recently, I've been moving back to joins, because MySQL is now a heck of a lot better at handling them than it was when I first started using it.
These days, a simple join on a properly indexed set of tables is rarely a problem, in performance terms. And if it does give a performance hit, then the use of index hints often solves them.
This is discussed here by one of the MySQL development team:
http://jorgenloland.blogspot.co.uk/2013/02/dbt-3-q3-6-x-performance-in-mysql-5610.html
So the summary is: If you've been avoiding joins in the past because of MySQL's abysmal performance with them, then try again on the latest versions. You'll probably be pleasantly surprised.
6 Comments
JOIN algorithms used in RDBMS' is called nested loops. It fundamentally is an N+1 select under the hood. The only difference is the DB made an intelligent choice to use it based off statistics and indexes, rather than client code forcing it down that path categorically.We moved away from the ORM in Django because of this problem. Basically, if you try and do
for p in person:
print p.car.colour
The ORM will happily return all people (typically as instances of a Person object), but then it will need to query the car table for each Person.
A simple and very effective approach to this is something I call "fanfolding", which avoids the nonsensical idea that query results from a relational database should map back to the original tables from which the query is composed.
Step 1: Wide select
select * from people_car_colour; # this is a view or sql function
This will return something like
p.id | p.name | p.telno | car.id | car.type | car.colour
-----+--------+---------+--------+----------+-----------
2 | jones | 2145 | 77 | ford | red
2 | jones | 2145 | 1012 | toyota | blue
16 | ashby | 124 | 99 | bmw | yellow
Step 2: Objectify
Suck the results into a generic object creator with an argument to split after the third item. This means that "jones" object won't be made more than once.
Step 3: Render
for p in people:
print p.car.colour # no more car queries
See this web page for an implementation of fanfolding for python.
4 Comments
select_related, which is meant to solve this - in fact, its docs start with an example similar to your p.car.colour example.select_related() and prefetch_related() in Django now.select_related() and friend don't seem to do any of the obviously useful extrapolations of a join such as LEFT OUTER JOIN. The problem isn't an interface problem, but an issue to do with the strange idea that objects and relational data are mappable....in my view.Suppose you have COMPANY and EMPLOYEE. COMPANY has many EMPLOYEES (i.e. EMPLOYEE has a field COMPANY_ID).
In some O/R configurations, when you have a mapped Company object and go to access its Employee objects, the O/R tool will do one select for every employee, wheras if you were just doing things in straight SQL, you could select * from employees where company_id = XX. Thus N (# of employees) plus 1 (company)
This is how the initial versions of EJB Entity Beans worked. I believe things like Hibernate have done away with this, but I'm not too sure. Most tools usually include info as to their strategy for mapping.
1 Comment
Here's a good description of the problem
Now that you understand the problem it can typically be avoided by doing a join fetch in your query. This basically forces the fetch of the lazy loaded object so the data is retrieved in one query instead of n+1 queries. Hope this helps.
Comments
Check Ayende post on the topic: Combating the Select N + 1 Problem In NHibernate.
Basically, when using an ORM like NHibernate or EntityFramework, if you have a one-to-many (master-detail) relationship, and want to list all the details per each master record, you have to make N + 1 query calls to the database, "N" being the number of master records: 1 query to get all the master records, and N queries, one per master record, to get all the details per master record.
More database query calls → more latency time → decreased application/database performance.
However, ORMs have options to avoid this problem, mainly using JOINs.
1 Comment
In my opinion the article written in Hibernate Pitfall: Why Relationships Should Be Lazy is exactly opposite of real N+1 issue is.
If you need correct explanation please refer Hibernate - Chapter 19: Improving Performance - Fetching Strategies
Select fetching (the default) is extremely vulnerable to N+1 selects problems, so we might want to enable join fetching
2 Comments
The supplied link has a very simply example of the n + 1 problem. If you apply it to Hibernate it's basically talking about the same thing. When you query for an object, the entity is loaded but any associations (unless configured otherwise) will be lazy loaded. Hence one query for the root objects and another query to load the associations for each of these. 100 objects returned means one initial query and then 100 additional queries to get the association for each, n + 1.
Comments
N+1 select issue is a pain, and it makes sense to detect such cases in unit tests. I have developed a small library for verifying the number of queries executed by a given test method or just an arbitrary block of code - JDBC Sniffer
Just add a special JUnit rule to your test class and place annotation with expected number of queries on your test methods:
@Rule
public final QueryCounter queryCounter = new QueryCounter();
@Expectation(atMost = 3)
@Test
public void testInvokingDatabase() {
// your JDBC or JPA code
}
Comments
The issue, as others have stated more elegantly, is that you either have a Cartesian product of the OneToMany columns or you're doing N+1 Selects. Either possible gigantic resultset or chatty with the database, respectively.
I'm surprised this isn't mentioned but this how I have gotten around this issue... I make a semi-temporary ids table. I also do this when you have the IN () clause limitation.
This doesn't work for all cases (probably not even a majority) but it works particularly well if you have a lot of child objects such that the Cartesian product will get out of hand (ie lots of OneToMany columns the number of results will be a multiplication of the columns) and its more of a batch like job.
First you insert your parent object ids as batch into an ids table. This batch_id is something we generate in our app and hold onto.
INSERT INTO temp_ids
(product_id, batch_id)
(SELECT p.product_id, ?
FROM product p ORDER BY p.product_id
LIMIT ? OFFSET ?);
Now for each OneToMany column you just do a SELECT on the ids table INNER JOINing the child table with a WHERE batch_id= (or vice versa). You just want to make sure you order by the id column as it will make merging result columns easier (otherwise you will need a HashMap/Table for the entire result set which may not be that bad).
Then you just periodically clean the ids table.
This also works particularly well if the user selects say 100 or so distinct items for some sort of bulk processing. Put the 100 distinct ids in the temporary table.
Now the number of queries you are doing is by the number of OneToMany columns.
Comments
Without going into tech stack implementation details, architecturally speaking there are at least two solutions to N + 1 Problem:
- Have Only 1 - big query - with Joins. This makes a lot of information be transported from the database to the application layer, especially if there are multiple child records. The typical result of a database is a set of rows, not graph of objects (there are solutions to that with different DB systems)
- Have Two(or more for more children needed to be joined) Queries - 1 for the parent and after you have them - query by IDs the children and map them. This will minimize data transfer between the DB and APP layers.
1 Comment
A generalisation of N+1
The N+1 problem is an ORM specific name of a problem where you move loops that could be reasonably executed on a server to the client. The generic problem isn't specific to ORMs, you can have it with any remote API. In this article, I've shown how JDBC roundtrips are very costly, if you're calling an API N times instead of only 1 time. The difference in the example is whether you're calling the Oracle PL/SQL procedure:
dbms_output.get_lines(call it once, receive N items)dbms_output.get_line(call it N times, receive 1 item each time)
They're logically equivalent, but due to the latency between server and client, you're adding N latency waits to your loop, instead of waiting only once.
The ORM case
In fact, the ORM-y N+1 problem isn't even ORM specific either, you can achieve it by running your own queries manually as well, e.g. when you do something like this in PL/SQL:
-- This loop is executed once
for parent in (select * from parent) loop
-- This loop is executed N times
for child in (select * from child where parent_id = parent.id) loop
...
end loop;
end loop;
It would be much better to implement this using a join (in this case):
for rec in (
select *
from parent p
join child c on c.parent_id = p.id
)
loop
...
end loop;
Now, the loop is executed only once, and the logic of the loop has been moved from the client (PL/SQL) to the server (SQL), which can even optimise it differently, e.g. by running a hash join (O(N)) rather than a nested loop join (O(N log N) with index)
Auto-detecting N+1 problems
If you're using JDBC, you could use jOOQ as a JDBC proxy behind the scenes to auto-detect your N+1 problems. jOOQ's parser normalises your SQL queries and caches data about consecutive executions of parent and child queries. This even works if your queries aren't exactly the same, but semantically equivalent.
Comments
ORM "N plus one" Problem
The "N plus one" problem is a common performance issue that can occur when using Object-Relational Mapping (ORM) frameworks. ORM frameworks are tools used to map database tables to objects in object-oriented programming languages. This problem arises when retrieving data from a relational database using ORM in a specific way.
To understand the "N plus one" problem, let's consider an example scenario where
you have two tables: Customer and Order. Each customer can have multiple
orders, and there is a one-to-many relationship between the Customer and
Order tables. In ORM, you define these relationships using object-oriented
concepts such as classes and references.
Now, let's say you want to retrieve all the customers along with their orders. In ORM, you might use a query like this:
customers = Customer.objects.all()
for customer in customers:
orders = customer.orders.all()
# Do something with the orders
In this code, you first retrieve all the customers using
Customer.objects.all(). Then, for each customer, you retrieve their orders
using customer.orders.all().
The issue with this approach is that it results in multiple queries being executed to the database. For example, if you have 100 customers, this code will execute 101 queries: one to retrieve all the customers and 100 more to retrieve the orders for each customer (hence the name "N plus one" problem). This can significantly impact performance, especially when dealing with large datasets.
The "N plus one" problem arises because the ORM framework performs a separate query for each customer's orders instead of fetching all the necessary data in a single query. This behavior is often the default in ORM frameworks to avoid unnecessarily loading all the associated data, which can be a performance concern in other scenarios.
To mitigate the "N plus one" problem, ORM frameworks usually provide ways to optimize data retrieval, such as eager loading or explicit joins. Eager loading allows you to fetch the required data in a single query, reducing the number of database round-trips. By specifying the relationships you want to include, the ORM framework can generate a more efficient query that retrieves all the necessary data at once.
As a demonstration of the "N plus one" problem and its solution, the following shows the actual SQL emitted from an ORM using SQLAlchemy.
Original ORM query with the N plus one problem (1 query for customers and N for each customer's order):
with Session(engine) as session:
customers = session.scalars(select(Customer))
for customer in customers:
print(f"> Customer: #{customer.customer_id}")
for order in customer.orders:
print(f"> order #{order.order_id} at {order.order_datetime}")
-- This query gets all customers:
SELECT customer.customer_id, ...
FROM customer
-- The following SQL is executed once for each customer:
SELECT "order".order_id AS order_order_id, ...
FROM "order"
WHERE "order".customer_id = %(param_1)s
After specifying eager loading (with selectinload()), only 2 queries are
required:
with Session(engine) as session:
customers = session.scalars(
select(Customer).options(selectinload(Customer.orders)))
for customer in customers:
print(f"> Customer: #{customer.customer_id}")
for order in customer.orders:
print(f"> order #{order.order_id} at {order.order_datetime}")
SELECT customer.customer_id, ...
FROM customer
-- This loads all the orders you need in one query:
SELECT "order".order_id AS order_order_id, ...
FROM "order"
WHERE "order".customer_id IN (%(primary_keys_1)s, %(primary_keys_2)s, ...)
Or, explicitly join and query the required fields (only 1 query is required):
with Session(engine) as session:
stmt = (
select(
Customer.customer_id,
Order.order_id,
Order.order_datetime,
)
.select_from(Customer)
.join(Customer.orders)
.order_by(Customer.customer_id)
)
results = session.execute(stmt)
current_customer_id = None
for row in results:
customer_id = row.customer_id
if current_customer_id != customer_id:
current_customer_id = customer_id
print(f"> Customer: #{current_customer_id}")
print(f"> order #{row.order_id} at {row.order_datetime}")
SELECT customer.customer_id, "order".order_id, ...
FROM customer
JOIN "order" ON customer.customer_id = "order".customer_id
ORDER BY customer.customer_id
In summary, the "N plus one" problem in ORM occurs when the framework executes multiple queries to retrieve associated data for each item in a collection, resulting in a significant performance overhead. Understanding and addressing this problem by optimizing data retrieval strategies can help improve the efficiency of ORM-based applications.
Comments
Take Matt Solnit example, imagine that you define an association between Car and Wheels as LAZY and you need some Wheels fields. This means that after the first select, hibernate is going to do "Select * from Wheels where car_id = :id" FOR EACH Car.
This makes the first select and more 1 select by each N car, that's why it's called n+1 problem.
To avoid this, make the association fetch as eager, so that hibernate loads data with a join.
But attention, if many times you don't access associated Wheels, it's better to keep it LAZY or change fetch type with Criteria.
1 Comment
N+1 SELECT problem is really hard to spot, especially in projects with large domain, to the moment when it starts degrading the performance. Even if the problem is fixed i.e. by adding eager loading, a further development may break the solution and/or introduce N+1 SELECT problem again in other places.
I've created open source library jplusone to address those problems in JPA based Spring Boot Java applications. The library provides two major features:
- Generates reports correlating SQL statements with executions of JPA operations which triggered them and places in source code of your application which were involved in it
2020年10月22日 18:41:43.236 DEBUG 14913 --- [ main] c.a.j.core.report.ReportGenerator : ROOT com.adgadev.jplusone.test.domain.bookshop.BookshopControllerTest.shouldGetBookDetailsLazily(BookshopControllerTest.java:65) com.adgadev.jplusone.test.domain.bookshop.BookshopController.getSampleBookUsingLazyLoading(BookshopController.java:31) com.adgadev.jplusone.test.domain.bookshop.BookshopService.getSampleBookDetailsUsingLazyLoading [PROXY] SESSION BOUNDARY OPERATION [IMPLICIT] com.adgadev.jplusone.test.domain.bookshop.BookshopService.getSampleBookDetailsUsingLazyLoading(BookshopService.java:35) com.adgadev.jplusone.test.domain.bookshop.Author.getName [PROXY] com.adgadev.jplusone.test.domain.bookshop.Author [FETCHING ENTITY] STATEMENT [READ] select [...] from author author0_ left outer join genre genre1_ on author0_.genre_id=genre1_.id where author0_.id=1 OPERATION [IMPLICIT] com.adgadev.jplusone.test.domain.bookshop.BookshopService.getSampleBookDetailsUsingLazyLoading(BookshopService.java:36) com.adgadev.jplusone.test.domain.bookshop.Author.countWrittenBooks(Author.java:53) com.adgadev.jplusone.test.domain.bookshop.Author.books [FETCHING COLLECTION] STATEMENT [READ] select [...] from book books0_ where books0_.author_id=1
- Provides API which allows to write tests checking how effectively your application is using JPA (i.e. assert amount of lazy loading operations )
@SpringBootTest
class LazyLoadingTest {
@Autowired
private JPlusOneAssertionContext assertionContext;
@Autowired
private SampleService sampleService;
@Test
public void shouldBusinessCheckOperationAgainstJPlusOneAssertionRule() {
JPlusOneAssertionRule rule = JPlusOneAssertionRule
.within().lastSession()
.shouldBe().noImplicitOperations().exceptAnyOf(exclusions -> exclusions
.loadingEntity(Author.class).times(atMost(2))
.loadingCollection(Author.class, "books")
);
// trigger business operation which you wish to be asserted against the rule,
// i.e. calling a service or sending request to your API controller
sampleService.executeBusinessOperation();
rule.check(assertionContext);
}
}
com.yannbriancon:spring-hibernate-query-utilslibrary.