What is @BatchSize and how does it solve the N+1 problem differently from JOIN FETCH?

@Entity
public class Customer {
    @Id private Long id;
    private String name;

    @OneToMany(mappedBy="customer", fetch=FetchType.LAZY)
    @BatchSize(size=25)  // load 25 collections at once
    private List<Order> orders;
}

// Without @BatchSize:
// 1 query: SELECT * FROM customers LIMIT 100
// 100 queries: SELECT * FROM orders WHERE customer_id=1 ... customer_id=100
// Total: 101 queries

// With @BatchSize(size=25):
// 1 query: SELECT * FROM customers LIMIT 100
// 4 queries: SELECT * FROM orders WHERE customer_id IN (1..25)
//             SELECT * FROM orders WHERE customer_id IN (26..50)
//             SELECT * FROM orders WHERE customer_id IN (51..75)
//             SELECT * FROM orders WHERE customer_id IN (76..100)
// Total: 5 queries!

// Global default:
# spring.jpa.properties.hibernate.default_batch_fetch_size=100

// @BatchSize on entity for proxy loading:
@Entity @BatchSize(size=20)
public class Product { ... }