Why should not use the AUTO JPA GenerationType with MySQL and Hibernate

Imagine having a tool that can automatically detect JPA and Hibernate performance issues. Hypersistence Optimizer is that tool!

Introduction

As I already mentioned, you should never use the TABLE identifier generator since it does not scale properly. In this post, I’ll show you why you should not rely on the AUTO GenerationType strategy if you’re Hibernate application uses MySQL.

Prior to Hibernate 5

On Hibernate 4, if you had the following entity mapping:

@Entity(name = "Post")
@Table(name = "post")
public class Post {

    @Id
    @GeneratedValue(strategy = GenerationType.AUTO)
    private Long id;

    private String title;

    public Post() {}

    public Post(String title) {
        this.title = title;
    }
}

When persisting 3 Post entities:

for ( int i = 1; i <= 3; i++ ) {
    entityManager.persist( 
        new Post( 
            String.format( 
                "High-Performance Java Persistence, Part %d", i
            ) 
        ) 
    );
}

Hibernate would generate the following insert statements:

INSERT INTO post (title) 
VALUES ('High-Performance Java Persistence, Part 1')

INSERT INTO post (title) 
VALUES ('High-Performance Java Persistence, Part 2')

INSERT INTO post (title) 
VALUES ('High-Performance Java Persistence, Part 3')

That’s great! Hibernate used the IDENTITY column to generate the entity identifier which is the only reasonable option for MySQL.

Hibernate 5

If you run the same unit test on Hibernate 5, you’ll get the following SQL statements:

SELECT next_val as id_val 
FROM hibernate_sequence FOR UPDATE

UPDATE hibernate_sequence 
SET next_val= 2 where next_val=1

SELECT next_val as id_val 
FROM hibernate_sequence FOR UPDATE

UPDATE hibernate_sequence 
SET next_val= 3 where next_val=1

SELECT next_val as id_val 
FROM hibernate_sequence FOR UPDATE

UPDATE hibernate_sequence 
SET next_val= 4 where next_val=3

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)

What’s just happened? Well, Hibernate picks the TABLE generator instead of IDENTITY when the underlying database does not support sequences. However, TABLE generator is not a good choice. Check out the HHH-11014 Jira issue for more details related to this behavior change.

How to fix it?

The fix is extremely easy. You just need to use the native identifier instead:

@Id
@GeneratedValue(
    strategy= GenerationType.AUTO, 
    generator="native"
)
@GenericGenerator(
    name = "native", 
    strategy = "native"
)
private Long id;

Now, when running the previous test case, Hibernate uses the IDENTITY column instead:

INSERT INTO post (title) 
VALUES ('High-Performance Java Persistence, Part 1')

INSERT INTO post (title) 
VALUES ('High-Performance Java Persistence, Part 2')

INSERT INTO post (title) 
VALUES ('High-Performance Java Persistence, Part 3')

If you want to use a portable solution that manages to customize the SEQUENCE generator while still allowing you to pick the IDENTITY generator for MySQL, then check out this article.

Online Workshops

If you enjoyed this article, I bet you are going to love my upcoming 4-day Online Workshop!

Conclusion

JPA portability is a myth! In reality, you have to know the details of the underlying JPA provider if you want a high-performance enterprise application.

Transactions and Concurrency Control eBook

Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.

4 day training with a Java Champion 🏆 29th Nov - 2nd Dec