[JPA_Basic] Fetch Join

이 글은 김영한의 [자바 ORM 표준 JPA 프로그래밍 - 기본편]을 수강하며 정리한 글입니다.

 

👉 기본 환경

- Language: Java

- DB: H2 Database

- IDE: IntelliJ

 

 

Fetch Join

    - JPQL에서 성능 최적화를 위해 제공하는 기능

    - 연관된 Entity나 Collection을 SQL 한 번으로 함께 조회하는 기능

-- JPQL
select m from Member m join fetch m.team;
 
-- SQL
SELECT M.*, T.* FROM MEMBER M INNER JOIN TEAM T ON M.TEAM_ID=T.ID;
 

 

Join

⌨️ 코드

public class Main {
    public static void main(String[] args) {
 
        // 생략
 
        try {
            Team teamA = new Team();
            teamA.setName("TeamA");
            em.persist(teamA);
 
            Team teamB = new Team();
            teamB.setName("TeamB");
            em.persist(teamB);
 
 
            Member memberA = new Member();
            memberA.setName("MemberA");
            memberA.setTeam(teamA);
            em.persist(memberA);
 
            Member memberB = new Member();
            memberB.setName("memberB");
            memberB.setTeam(teamA);
            em.persist(memberB);
 
            Member memberC = new Member();
            memberC.setName("memberC");
            memberC.setTeam(teamB);
            em.persist(memberC);
 
            em.flush();
            em.clear();
 
            String query = "select m from Member m ";
 
            List<Member> members = em.createQuery(query, Member.class)
                    .getResultList();
 
            for (Member m : members)
                System.out.println("member: " + m);
 
            tx.commit(); // transaction 종료 후 commit
        } catch (Exception e) {
            tx.rollback(); // 문제가 생길 경우, rollback 진행
            e.printStackTrace();
        } finally {
            em.close(); // tx에 문제가 생기더라도 em 반드시 종료
        }
 
 
        emf.close();
        // 트랜잭션 단위로 관리되는 Entity Manager는 Tx가 종료되면 close가 되어야하지만, emf는 Application 종료 시, close되어야 함
 
    }
}
 

 

🖨️발생한 쿼리

Hibernate: 
    /* select
        m 
    from
        Member m  */ select
            member0_.id as id1_0_,
            member0_.age as age2_0_,
            member0_.name as name3_0_,
            member0_.TEAM_ID as team_id5_0_,
            member0_.type as type4_0_ 
        from
            Member member0_
 

@ManyToOne 관계로 fetchType.LAZY 설정

    - Team이 사용되지 않으면 select이 실행되지 않음

일반 조인 실행 시, 연관된 엔티티를 함께 조회하지 않음

 

⌨️ 코드

public class Main {
    public static void main(String[] args) {
 
        // 생략
 
        try {
 
            // 생략
 
            String query = "select m from Member m ";
 
            List<Member> members = em.createQuery(query, Member.class)
                    .getResultList();
 
            for (Member m : members)
                System.out.println("member: " + m + ", member.team" + m.getTeam().getName());
 
            tx.commit(); // transaction 종료 후 commit
        } catch (Exception e) {
            tx.rollback(); // 문제가 생길 경우, rollback 진행
            e.printStackTrace();
        } finally {
            em.close(); // tx에 문제가 생기더라도 em 반드시 종료
        }
 
 
        emf.close();
        // 트랜잭션 단위로 관리되는 Entity Manager는 Tx가 종료되면 close가 되어야하지만, emf는 Application 종료 시, close되어야 함
 
    }
}
 

 

🖨️발생한 쿼리

Hibernate: 
    /* select
        m 
    from
        Member m  */ select
            member0_.id as id1_0_,
            member0_.age as age2_0_,
            member0_.name as name3_0_,
            member0_.TEAM_ID as team_id5_0_,
            member0_.type as type4_0_ 
        from
            Member member0_
Hibernate: 
    select
        team0_.id as id1_3_0_,
        team0_.name as name2_3_0_ 
    from
        Team team0_ 
    where
        team0_.id=?
member: Member{id=3, name='MemberA', age=0}, member.teamTeamA
member: Member{id=4, name='memberB', age=0}, member.teamTeamA
Hibernate: 
    select
        team0_.id as id1_3_0_,
        team0_.name as name2_3_0_ 
    from
        Team team0_ 
    where
        team0_.id=?
member: Member{id=5, name='memberC', age=0}, member.teamTeamB
 

🚨 N+1 문제 발생

    - Member 객체에 포함된 Team 개수만큼 select query 발생

        - SQL: MemberA + TeamA

        - 1차 캐시: MemberB + TeamA

        - SQL: MemberC + TeamB

 

 

Fetch Join

⌨️ 코드

public class Main {
    public static void main(String[] args) {
 
        // 생략
 
        try {
 
            // 생략
 
            String query = "select m from Member m join fetch m.team ";
 
            List<Member> members = em.createQuery(query, Member.class)
                    .getResultList();
 
            for (Member m : members)
                System.out.println("member: " + m + ", member.team" + m.getTeam().getName());
 
            tx.commit(); // transaction 종료 후 commit
        } catch (Exception e) {
            tx.rollback(); // 문제가 생길 경우, rollback 진행
            e.printStackTrace();
        } finally {
            em.close(); // tx에 문제가 생기더라도 em 반드시 종료
        }
 
 
        emf.close();
        // 트랜잭션 단위로 관리되는 Entity Manager는 Tx가 종료되면 close가 되어야하지만, emf는 Application 종료 시, close되어야 함
 
    }
}
 

 

🖨️발생한 쿼리

Hibernate: 
    /* select
        m 
    from
        Member m 
    join
        fetch m.team  */ select
            member0_.id as id1_0_0_,
            team1_.id as id1_3_1_,
            member0_.age as age2_0_0_,
            member0_.name as name3_0_0_,
            member0_.TEAM_ID as team_id5_0_0_,
            member0_.type as type4_0_0_,
            team1_.name as name2_3_1_ 
        from
            Member member0_ 
        inner join
            Team team1_ 
                on member0_.TEAM_ID=team1_.id
member: Member{id=3, name='MemberA', age=0}, member.teamTeamA
member: Member{id=4, name='memberB', age=0}, member.teamTeamA
member: Member{id=5, name='memberC', age=0}, member.teamTeamB
 

Fetch Join으로 회원과 팀을 함께 조회하여 지연 로딩 X

 

 

Collection Fetch Join

⌨️ 코드

public class Main {
    public static void main(String[] args) {
 
        // 생략
 
        try {
 
            // 생략
 
            String query = "select t from Team t join fetch t.members ";
 
            List<Team> teams = em.createQuery(query, Team.class)
                    .getResultList();
 
            for (Team t : teams){
                System.out.println("Team: " + t.getName());
                for(Member m: t.getMembers()){
                    System.out.println(" -> Member: " + m);
                }
            }
 
            tx.commit(); // transaction 종료 후 commit
        } catch (Exception e) {
            tx.rollback(); // 문제가 생길 경우, rollback 진행
            e.printStackTrace();
        } finally {
            em.close(); // tx에 문제가 생기더라도 em 반드시 종료
        }
 
 
        emf.close();
        // 트랜잭션 단위로 관리되는 Entity Manager는 Tx가 종료되면 close가 되어야하지만, emf는 Application 종료 시, close되어야 함
 
    }
}
 

 

🖨️발생한 쿼리

Hibernate: 
    /* select
        t 
    from
        Team t 
    join
        fetch t.members  */ select
            team0_.id as id1_3_0_,
            members1_.id as id1_0_1_,
            team0_.name as name2_3_0_,
            members1_.age as age2_0_1_,
            members1_.name as name3_0_1_,
            members1_.TEAM_ID as team_id5_0_1_,
            members1_.type as type4_0_1_,
            members1_.TEAM_ID as team_id5_0_0__,
            members1_.id as id1_0_0__ 
        from
            Team team0_ 
        inner join
            Member members1_ 
                on team0_.id=members1_.TEAM_ID
Team: TeamA
 -> Member: Member{id=3, name='MemberA', age=0}
 -> Member: Member{id=4, name='memberB', age=0}
Team: TeamA
 -> Member: Member{id=3, name='MemberA', age=0}
 -> Member: Member{id=4, name='memberB', age=0}
Team: TeamB
 -> Member: Member{id=5, name='memberC', age=0}
 

🚨 record 수에 맞춰서 Team이 중복되어 출력됨

 

Distinct

    - SQL의 DISTINCT 기능(데이터가 다르면 의도한 DISTINCT가 제대로 동작하지 X)

    - 애플리케이션 엔티티 중복 제거(같은 식별자를 가진 엔티티 제거)

 

⌨️ 코드

public class Main {
    public static void main(String[] args) {
 
        // 생략
 
        try {
 
            // 생략
 
            String query = "select distinct t from Team t join fetch t.members ";
 
            List<Team> teams = em.createQuery(query, Team.class)
                    .getResultList();
 
            for (Team t : teams){
                System.out.println("Team: " + t.getName());
                for(Member m: t.getMembers()){
                    System.out.println(" -> Member: " + m);
                }
            }
 
            tx.commit(); // transaction 종료 후 commit
        } catch (Exception e) {
            tx.rollback(); // 문제가 생길 경우, rollback 진행
            e.printStackTrace();
        } finally {
            em.close(); // tx에 문제가 생기더라도 em 반드시 종료
        }
 
 
        emf.close();
        // 트랜잭션 단위로 관리되는 Entity Manager는 Tx가 종료되면 close가 되어야하지만, emf는 Application 종료 시, close되어야 함
 
    }
}
 

 

🖨️발생한 쿼리

Hibernate: 
    /* select
        distinct t 
    from
        Team t 
    join
        fetch t.members  */ select
            distinct team0_.id as id1_3_0_,
            members1_.id as id1_0_1_,
            team0_.name as name2_3_0_,
            members1_.age as age2_0_1_,
            members1_.name as name3_0_1_,
            members1_.TEAM_ID as team_id5_0_1_,
            members1_.type as type4_0_1_,
            members1_.TEAM_ID as team_id5_0_0__,
            members1_.id as id1_0_0__ 
        from
            Team team0_ 
        inner join
            Member members1_ 
                on team0_.id=members1_.TEAM_ID
Team: TeamA
 -> Member: Member{id=3, name='MemberA', age=0}
 -> Member: Member{id=4, name='memberB', age=0}
Team: TeamB
 -> Member: Member{id=5, name='memberC', age=0}
 

컬렉션 내 중복 제거

 

🤓 member1_TEAM_ID 가 select절에 2번 쓰인 이유

    - 하나의 별칭(team_id5_0_0__)은 Team 객체를 구성하는데 사용

    - 다른 하나의 별칭(team_id5_0_1_)은 Member 객체를 구성하는데 사용
    - 같은 필드에 대해 서로 다른 용도로 별칭을 부여함으로써, 하나의 SQL 쿼리 결과로 여러 개의 객체를 올바르게 생성할 수 있음

따라서 실제 데이터베이스에서 members1_.TEAM_ID 필드가 두 번 조회되는 것은 아니며, 하이버네이트가 내부적으로 같은 필드를 서로 다른 용도로 활용하기 위해 별칭을 두 번 부여한 것