More Books
JBoss 4.0 The Official Guide
JBoss® 4.0 The Official Guide
Table of Contents
Copyright
About the Authors
We Want to Hear from You!
Introduction
What This Book Covers
About JBoss
About Open Source
About Professional Open Source
What's New in JBoss 4.0
Chapter 1.  Installing and Building the JBoss Server
Getting the Binary Files
Installing the Binary Package
Basic Installation Testing
Booting from a Network Server
Building the Server from Source Code
Chapter 2.  The JBoss JMX Microkernel
JMX
The JBoss JMX Implementation Architecture
Connecting to the JMX Server
Using JMX as a Microkernel
The JBoss Deployer Architecture
Exposing MBean Events via SNMP
Remote Access to Services, Detached Invokers
Chapter 3.  Naming on JBoss
An Overview of JNDI
The JBossNS Architecture
Chapter 4.  Transactions on JBoss
Transaction and JTA Overview
JBoss Transaction Internals
Chapter 5.  EJBs on JBoss
The EJB Client-Side View
The EJB Server-Side View
The EJB Container
Entity Bean Locking and Deadlock Detection
Chapter 6.  Messaging on JBoss
JMS Examples
JBossMQ Overview
JBossMQ Configuration and MBeans
Specifying the MDB JMS Provider
Chapter 7.  Connectors on JBoss
JCA Overview
An Overview of the JBossCX Architecture
Configuring JDBC Datasources
Configuring Generic JCA Adaptors
Chapter 8.  Security on JBoss
J2EE Declarative Security Overview
An Introduction to JAAS
The JBoss Security Model
The JBossSX Architecture
The Secure Remote Password (SRP) Protocol
Running JBoss with a Java 2 Security Manager
Using SSL with JBoss and JSSE
Configuring JBoss for Use Behind a Firewall
Securing the JBoss Server
Chapter 9.  Web Applications
The Tomcat Service
The Tomcat server.xml File
The Engine Element
The Host Element
Using SSL with the JBoss/Tomcat Bundle
Setting the Context Root of a Web Application
Setting Up Virtual Hosts
Serving Static Content
Using Apache with Tomcat
Using Clustering
Integrating Third-Party Servlet Containers
Chapter 10.  MBean Services Miscellany
System Properties Management
Property Editor Management
Services Binding Management
Scheduling Tasks
The Log4j Service MBean
RMI Dynamic Class Loading
Chapter 11.  The CMP Engine
Example Code
The jbosscmp-jdbc Structure
Entity Beans
CMP Fields
Container-Managed Relationships
Declaring Queries
Optimized Loading
The Loading Process
Transactions
Optimistic Locking
Entity Commands and Primary Key Generation
JBoss Global Defaults
Datasource Customization
Chapter 12.  Web Services
JAX-RPC Service Endpoints
Enterprise JavaBean Endpoints
Web Services ClientsA JAX-RPC Client
Service References
Chapter 13.  Hibernate
The Hibernate MBean
Hibernate Archives
Using Hibernate Objects
Using a HAR File Inside an EAR File
The HAR Deployer
Chapter 14.  Aspect-Oriented Programming (AOP) Support
JBoss AOP: EJB-Style Services for Plain Java Objects
Why AOP?
Basic Concepts of AOP
Building JBoss AOP Applications
The JBoss AOP Deployer
Packaging and Deploying AOP Applications to JBoss
Appendix A.  The GNU Lesser General Public License (LGPL)
GNU General Public License
Appendix B.  Example Installation
Index
SYMBOL
A
B
C
D
E
F
G
H
I
J
K
L
M
N
O
P
Q
R
S
T
U
V
W
X

Transactions

All the examples presented in this chapter so far have been defined to run in a transaction. Transaction granularity is a dominating factor in optimized loading because transactions define the lifetime of preloaded data. If a transaction completes, commits, or rolls back, the data in the preload cache is lost. This can result in a severe negative performance impact.

Let's examine the performance impact of running without a transaction by looking at an example that uses an on-find optimized query that selects the first four gangsters (to keep the result set small) and is executed without a wrapper transaction. The sample code follows:

public String createGangsterHtmlTable_no_tx() throws FinderException
{
    StringBuffer table = new StringBuffer();
    table.append("<table>");

    Collection gangsters = gangsterHome.findFour();
    for(Iterator iter = gangsters.iterator(); iter.hasNext(); ) {
        Gangster gangster = (Gangster)iter.next();
        table.append("<tr>");
        table.append("<td>").append(gangster.getName());
        table.append("</td>");
        table.append("<td>").append(gangster.getNickName());
        table.append("</td>");
        table.append("<td>").append(gangster.getBadness());
        table.append("</td>");
        table.append("</tr>");
    }

    table.append("</table>");
    return table.toString();
}

The finder results in the following query being executed:

SELECT t0_g.id, t0_g.name, t0_g.nick_name, t0_g.badness
  FROM gangster t0_g
  WHERE t0_g.id < 4
  ORDER BY t0_g.id ASC

Normally, this would be the only query executed, but because this code is not running in a transaction, all the preloaded data is thrown away as soon as the finder returns. Then when the CMP field is accessed, JBoss executes the following four queries (one for each loop):

SELECT id, name, nick_name, badness
  FROM gangster
  WHERE (id=0) OR (id=1) OR (id=2) OR (id=3)
SELECT id, name, nick_name, badness
  FROM gangster
  WHERE (id=1) OR (id=2) OR (id=3)
SELECT id, name, nick_name, badness
  FROM gangster
  WHERE (id=2) OR (id=3)
SELECT name, nick_name, badness
  FROM gangster
  WHERE (id=3)

It's actually worse than this. JBoss executes each of these queries three times, once for each CMP field that is accessed. This is because the preloaded values are discarded between the CMP field accessor calls. Figure 11.13 shows the execution of the queries.

Figure 11.13. No transaction on-find optimized query execution.


This performance is much worse than that of read-ahead none because of the amount of data loaded from the database. The number of rows loaded is determined by the following equation:


This all happens because the transaction in the example is bounded by a single call on the entity. This brings up the important question "How do I run my code in a transaction?" The answer depends on where the code runs. If it runs in an EJB (session, entity, or message driven), the method must be marked with the Required or RequiresNew trans-attribute in the assembly-descriptor. If the code is not running in an EJB, a user transaction is necessary. The following code wraps a call to the declared method with a user transaction:

public String createGangsterHtmlTable_with_tx()
    throws FinderException
{
    UserTransaction tx = null;
    try {
        InitialContext ctx = new InitialContext();
        tx = (UserTransaction) ctx.lookup("UserTransaction");
        tx.begin();
        
        String table = createGangsterHtmlTable_no_tx();
       
        if (tx.getStatus() == Status.STATUS_ACTIVE) {
                tx.commit();
        }
            return table;
    } catch (Exception e) {
        try {
            if (tx != null) tx.rollback();
        } catch (SystemException unused) {
            // eat the exception we are exceptioning out anyway
        }
        if (e instanceof FinderException) {
                throw (FinderException) e;
        }
        if (e instanceof RuntimeException) {
                throw (RuntimeException) e;
        }
    
        throw new EJBException(e);
    }
}