Hypersistence Optimizer is a dynamic analyzing tool that can scan your JPA and Hibernate application and provide you tips about the changes you need to make to entity mappings, configurations, queries, and Persistence Context actions to speed up your data access layer.
Once you downloaded the Full or Trial version, you need to follow a series of steps in order to install Hypersistence Optimizer.
The first thing you need to do is to unzip the package you have just downloaded.
> unzip hypersistence-optimizer-2.10.1-pack.zip
After unzipping the project package, you will get the following file structure:
creating: hypersistence-optimizer-2.10.1/
creating: hypersistence-optimizer-2.10.1/lib/
inflating: hypersistence-optimizer-2.10.1/lib/hypersistence-optimizer-2.10.1-javadoc.jar
inflating: hypersistence-optimizer-2.10.1/lib/hypersistence-optimizer-2.10.1-sources.jar
inflating: hypersistence-optimizer-2.10.1/lib/hypersistence-optimizer-2.10.1.jar
creating: hypersistence-optimizer-2.10.1/docs/
creating: hypersistence-optimizer-2.10.1/docs/html/
creating: hypersistence-optimizer-2.10.1/docs/pdf/
inflating: hypersistence-optimizer-2.10.1/docs/html/asciidoctor.css
inflating: hypersistence-optimizer-2.10.1/docs/html/coderay-asciidoctor.css
inflating: hypersistence-optimizer-2.10.1/docs/pdf/InstallationGuide.pdf
inflating: hypersistence-optimizer-2.10.1/docs/html/InstallationGuide.html
inflating: hypersistence-optimizer-2.10.1/docs/html/UserGuide.html
inflating: hypersistence-optimizer-2.10.1/docs/pdf/UserGuide.pdf
inflating: hypersistence-optimizer-2.10.1/changelog.txt
inflating: hypersistence-optimizer-2.10.1/LICENSE.txt
inflating: hypersistence-optimizer-2.10.1/maven-install.bat
inflating: hypersistence-optimizer-2.10.1/maven-install.sh
inflating: hypersistence-optimizer-2.10.1/README.txt
The package contains the following resources:
- the
lib
folder contains the mainjar
file, as well as the JavaDoc and the Java sources - the
docs
folder contains the Installation and User Guides - the
changelog
file contains the release notes for all product versions - the
LICENSE
file contains the project license info - the
maven-install
scripts allow you to install the Java artifacts in your local Maven repository - the
README
file contains a short description of the project
In order to install Hypersistence Optimizer, you need to read the Installation Guide, which is available both in HTML and PDF formats in the unzipped package:
hypersistence-optimizer-2.10.1/docs/pdf/InstallationGuide.pdf
hypersistence-optimizer-2.10.1/docs/html/InstallationGuide.html
You can also read the Installation Guide online if you want.
After you are done with the Installation Guide, you should read the User Guide too, as it shows how you can configure Hypersistence Optimizer so that you can get the most out of it.
You can find the User Guide in the docs folder as well:
hypersistence-optimizer-2.10.1/docs/html/UserGuide.html
hypersistence-optimizer-2.10.1/docs/pdf/UserGuide.pdf
You can also read the User Guide online if you want.
This repository contains examples for Spring Boot, Spring Framework, Java EE that can help you set up the tool.
By default, this repository is configured for the Full version.
If you want to use it with the Trial version, you should use the trial
branch instead:
> git checkout trial
You should check out the entire repository since the child modules use the versions defined in the parent module
pom.xml
.
There are multiple modules in this repository:
hypersistence-optimizer-test-case
hypersistence-optimizer-config-example
hypersistence-optimizer-spring-boot-example
hypersistence-optimizer-spring-jpa-example
hypersistence-optimizer-spring-jpa-hibernate4-example
hypersistence-optimizer-spring-jpa-hibernate3-example
hypersistence-optimizer-spring-hibernate-example
hypersistence-optimizer-spring-hibernate4-example
hypersistence-optimizer-spring-hibernate3-example
hypersistence-optimizer-glassfish-hibernate-example
hypersistence-optimizer-glassfish-hibernate4-example
hypersistence-optimizer-quarkus-spring-extension-example
hypersistence-optimizer-quarkus-example
This module provides a test case template you could use to replicate a certain Hypersistence Optimizer issue.
This module shows various examples of how you can configure Hypersistence Optimizer.
This module shows how you can use Hypersistence Optimizer with Spring Boot.
This module shows how you can use Hypersistence Optimizer with Spring and the JPA LocalEntityManagerFactoryBean
.
This module shows how you can use Hypersistence Optimizer with Spring, the JPA LocalEntityManagerFactoryBean
, and Hibernate 4.
This module shows how you can use Hypersistence Optimizer with Spring, the JPA LocalEntityManagerFactoryBean
, and Hibernate 3.
This module shows how you can use Hypersistence Optimizer with Spring and the Hibernate LocalSessionFactoryBean
.
This module shows how you can use Hypersistence Optimizer with Spring and the Hibernate 4 LocalSessionFactoryBean
.
This module shows how you can use Hypersistence Optimizer with Spring and the Hibernate 3 LocalSessionFactoryBean
.
This module shows how you can use Hypersistence Optimizer with Java EE and GlassFish.
This module shows how you can use Hypersistence Optimizer with Java EE, GlassFish, and Hibernate 4.
This module shows how you can use Hypersistence Optimizer with Quarkus's Spring based extensions.
This module shows how you can use Hypersistence Optimizer with Quarkus (Hibernate Panache ORM).
All modules work with Java 8 or newer versions, and the project is set up using Maven.
While most tests use the in-memory HSQLDB, some tests use MySQL, and, for this reason, you need to configure
the MySQL database connection properties in the MySQLDataSourceProvider
class.
By default, the project expects that you are using the following MySQL database connection properties:
- Database: high_performance_java_persistence
- Port: 3306
- UserName: mysql
- Password: admin
You can also use docker-compose to run MySQL server using the command:
docker-compose up
To run the tests, you can use:
mvn clean test
If you'd like to create a new issue, be it a feature request or simply reporting a bug, then you can use this GitHub repository issue list.
For bugs, it would be awesome if you provided a replicating test case as well. You can use the EagerFetchingManyToOneTest
as a template to create your new test case scenario.
When you are done, please send your test case as a Pull Request, and I'll take care of it.
Thank you for using Hypersistence Optimizer and stay tuned for more!