Shulou(Shulou.com)06/02 Report--

This article focuses on "what are the methods of Hibernate performance optimization", interested friends may wish to take a look. The method introduced in this paper is simple, fast and practical. Next let the editor to take you to learn "what are the methods of Hibernate performance optimization?"

1. Try to use many-to-one and avoid using individual one-to-many

two。 Flexible use of one-way one-to-many

3. Instead of one-to-one, use many-to-one instead of one-to-one

4. Configure object caching without using collection caching

5. One-to-many use Bag many-to-one use Set

6. Inheritance uses display polymorphic HQL:from object polymorphism= "exlicit" to avoid checking and dealing with all objects

7. Eliminate large tables and use secondary caching. We explain all of the above in detail.

One-to-many: using inverse=false (default), the association of objects is maintained by parent objects. In the case of inverse=true, the general user has a two-way many-to-many association, and the relationship is maintained by sub-objects. When adding sub-objects, you need to show: child.setParent (child) in order to improve performance, you should try to use two-way one-to-many inverse=true, and you should directly use Session to persist objects in the DAO interface of the MVC structure. Avoid associative relationships (this sentence is a bit confusing), and the correct use of secondary caching in single relationships can greatly improve query-based applications.

There are fewer many-to-one performance problems, but avoid the classic Number1 problem.

Associating through a primary key is equivalent to splitting a large table into a small table. (this is a key point that distinguishes object-oriented design from process-oriented design.)

Correct use of list, bag and set one-to-many:

◆ using list needs to maintain the Index Column field, cannot be used for two-way association, and must use inverse=false, which needs to be used carefully in some rare situations (basically not considered for use)

The semantics of ◆ bag/set are basically the same in one-to-many, so it is recommended to use bag many-to-one:

◆ bag is different from set. Bag allows repetitive insertion. It is recommended that set be used in large collection paging to use session.createFilter session.createFilter (parent.getChildren (), "), setFirstResult (0), setMaxResult (10)) to avoid Number1 reference (http://www.javaeye.com/post/266972).

In the many-to-one case, querying the child object will result in a query of the parent class of each subclass on the page, which needs to be avoided by using the following method: many-to-one fetch= "join | select" (this method may be problematic)

Inverse=true cannot maintain the collection cache (I still don't quite understand the collection cache and object cache)

OLTP-type web applications can scale at the cluster level, and the inevitable database bottleneck framework can reduce the pressure of accessing the database. Caching is an important criterion to measure whether a framework is excellent or not. Hibernate from the perspective of cache

◆ object cache, fine granularity, is for the level of the table, transparent access, because there is the advantage of not changing the code, so it is the magic weapon of ORM to improve performance

◆ Hibernate is the current ORM framework for cache performance.

◆ query cache

* Robbin also makes a summary of the frequent discussions on Hibernate vs iBatis:

For OLTP applications, using ORM framework and OLEB applications (not sure what application it is) * * using JDBC or other methods to deal with Hibernate tends to be fine-grained design, object-oriented, splitting large tables into multiple small tables, eliminating redundant fields, and improving performance through secondary cache.

IBatis tends to be coarse-grained, relation-oriented, merges tables as much as possible, and eliminates associations through Column redundancy, but iBatis has no effective caching means.

It can be said that our Hibernate performance optimization summary plays an important role for developers using Hibernate.

At this point, I believe you have a deeper understanding of "what are the methods of Hibernate performance optimization?" you might as well do it in practice. Here is the website, more related content can enter the relevant channels to inquire, follow us, continue to learn!

Welcome to subscribe "Shulou Technology Information " to get latest news, interesting things and hot topics in the IT industry, and controls the hottest and latest Internet news, technology news and IT industry trends.