Programming message services in Java
Asynchronous interactions seem to be the way of the future
Lately I seem to be involved increasingly in systems relying on asynchronous interactions for efficient and effective integration. In these cases, the asynchronous behaviour has been implemented using JMS (or Java Message Service) queues, with message driven beans, and with the underlying message server provided by application servers such as WebLogic, WebSphere and JBoss.
As a point to note, "asynchronous" here indicates that the initiating process (the client) returns immediately that a message is sent rather than waiting for the invoked process to complete before returning. This is usually desirable if the invoked process is long running or may require human intervention.
Although JMS has been around for some time, I’ve found that knowledge of how JMS works and its use with EJB3 Message Driven Beans is actually quite limited and I realise that I’ve never actually written a column about JMS before. This Java column therefore looks at what JMS is, how it works, and at the use of queues. It also provides an example of implementing a JSM client and an EJB3 based Message Driven Bean (MDB) to act as the consumer of JMS messages.
Java Message Service
The Java Message Service (more commonly referred to as JMS) is a standard Java Standard Edition API that can be used to access a variety of different message servers. Although the name might suggest otherwise, it’s not of itself a message service, but rather a generic interface to a message server.
In this sense, it’s similar to the JDBC and JNDI, in that it is trying to provide a standardised and unified interface to a variety of disparate vendor specific message servers. In the case of JMS, you use JMS to interface to message services such as IBM’s MQ Series message service, the JBoss MQ messaging system, or Oracle AQ queues etc.
The JMS API is defined within the javax.jms package. This package defines numerous interfaces, only two classes, and a number of exceptions. Before going into further detail about JMS we’ll look at the general concepts behind messaging services.
Sponsored: Hyper-scale data management