Java ee – Google 2007 JavaOne Advance Conference Guide User Manual

Page 33

technical sessions | track four : java EE |

Java EE

TS-4721 Implementing Java EE Applications, using Enterprise

JavaBeans (EJB) 3 Technology: real-World Tips, Tricks,
and New design Patterns

Edgar A Silva, Summa Technologies
Fabiane Nardon, Vidatis

The new Enterprise JavaBeans (EJB) 3 specification makes enterprise
applications development easier, cleaner, and faster. As with any other
new technology, however, there are always new tricks to be learned and
pitfalls to be avoided. This session presents the speakers’ experience with
building real-world applications using EJB 3 technology. The session uses
the lessons learned in this experience to show techniques, tools, tips, and
tricks to make good use of session-beans-, entity-beans-, and message-
driven-beans-compliant objects, discussing the problems and pitfalls
encountered. It presents a series of refactoring actions you can perform
to better use new features in EJB 3 technology, with real-world examples.
The session also presents a revision of the traditional EJB technology
design patterns, showing the role they played when release 3 was being
developed and presenting new design patterns that arose with this new
version. The goal of this session is to share the results of this experience
with other developers, discussing the outcome of using EJB 3 technology,
the benefits, the drawbacks, and the techniques for building a successful
EJB 3 technology-based application.

TS-4746 Hibernate Search: Googling Your Java Technology-Based

Persistent domain Model

Emmanuel Bernard, JBoss

query and search capabilities are often synonyms of traditional SqL
queries for most applications. Full-text search capabilities (think
Google search engine in applications), however, bring a much wider
expressiveness and freedom to queries (synonyms, approaching terms,
noise words, result by relevance, and so on) and make them order(s) of
magnitude faster than an equivalent (if even possible) SqL query.

Despite its clear advantage, this technology is not common in traditional
Java EE applications. Some solutions tie you to a specific database engine,
and some don’t, and the cost and complexity of bringing a full-text search
engine into a Java EE application is often not negligible, due to the

• Mismatch between the domain model and the full-text index model
• Cost of keeping the index up to date in terms of enterprise data changes

(RDBMS)

• Mismatch between the index querying process and the ORM query API

such as the Java Persistence API

Application developers tend to exclude such a technology from their day-
to-day applications.

Built on top of Apache Lucene, Hibernate Search (an open source project
under the Hibernate umbrella) nails down each of those issues, bringing a
Google-like search capability to persistent domain models powered by the
Java Persistence API. This session shows how full-text search features have
been integrated into the Java Persistence API and metadata, bringing a
unified view to your domain model and avoiding the previous mismatches.
Such ORM/search engine integration lowers the barrier to entry for full-
text technologies.

A live demo shows how to add full-text search capabilities to an existing
application based on the Enterprise JavaBeans (EJB) 3.0 specification.

After this session, you should know how to add full-text search engine
capabilities to existing code and reconsider the cost and complexity of
such an operation, opening new horizons to your business applications.

A general understanding of ORM concepts is expected for this session.

TS-4902 Java Persistence API: Best Practices and Tips

Mitesh Meswani, Sun Microsystems, Inc.
Rima Patel Sriganesh, Sun Microsystems, Inc.

It has been almost a year since the release of the Java Persistence API
specification, and many enterprise developers have started using this
technology in their Java EE and Java SE applications. As a consequence,
the recurring questions JPA developers have been posing need to
be addressed. This session goes through some of the best practices
for designing efficient JPA applications and presents techniques for
achieving functionality in JPA-based applications that is currently lacking
in the core API.

The session covers the following topics:

• Basics of high-performance JPA entities
- Flush mode
- Locking strategy
- Caching
• Whether to use a container-managed or application-managed entity

manager in web tier applications

• Generated primary keys—are they as simple as they sound?
• Scenarios regarding detach and merge
• Bulk updates and deletes—things to remember
• Impact of data model on entity performance
• Tips for mapping objects to related data
• Mapping overrides using XML
• Transactions and entities
- Transaction isolation levels
• Using stored procedures with JPA

TS-4593 Guidelines, Tips, and Tricks in Using Java EE 5 from the Java BluePrints Program

TS-4656 Harvard-MIT Data Center’s Dataverse Network: A JavaServer Faces/Enterprise

JavaBeans 3.0 Technology Digital Library Application on Java EE 5

TS-4721 Implementing Java EE Applications, Using Enterprise JavaBeans (EJB) 3

Technology: Real-World Tips, Tricks, and New Design Patterns

TS-4746 Hibernate Search: Googling Your Java Technology-Based Persistent

Domain Model

TS-4902 Java Persistence API: Best Practices and Tips

track four : Java EE