April 21, 2013 Leave a comment
Is your app running slow… how about making some part of your code asynchronous. Its really easy and 5 minutes task to make it so using Spring Task Execution API. Why to bother about creating your own thread objects, implementing interfaces.
Try this out, a 5 minutes recipe…..
Spring 3 provides a Task Execution API that allows us to wrap beans with a proxy that calls @Async annotated methods asynchronously. The easiest approach would be simply to annotate your desired service method with @Async, however you need to limit the scope of such method on the forked thread, which means no db operations, no user interaction, etc can go inside such method.
HOW TO CONFIGURE YOUR APP FOR ASYNCHRONOUS TASKS
Just adding the @Async annotation isn’t enough. We need to do little configuration in applicationContext.xml and all done…
Step 1: Add Task namespace and Task namespace schema location
<beans xmlns="http://www.springframework.org/schema/beans"</span> xmlns:task="http://www.springframework.org/schema/task" xmlns:p="http://www.springframework.org/schema/p" ....... xmlns:jbpm="http://drools.org/schema/drools-spring" xmlns:cache="http://www.springframework.org/schema/cache" xsi:schemaLocation="http://www.springframework.org/schema/jee ..... http://www.springframework.org/schema/task http://www.springframework.org/schema/task/spring-task-3.0.xsd">
Step2: Activate @Async and Task executor for asynchronous calls
<task:annotation-driven /> <task:executor id="ekramExecutor" pool-size="5-15" queue-capacity="50" keep-alive="500" />
Details of task:executor
|pool-size||Number of threads in the pool. Either a single integer or a range such as "5-15".|
|queue-capacity||Task queue capacity.|
|keep-alive||time in seconds for threads in excess of the core pool size.|
|rejection-policy||Policy when threads are rejected (queue exhausted and pool size at maximum). Options are ABORT (default), CALLER_RUNS, DISCARD and DISCARD_OLDEST.|
For complete project, check out https://github.com/ekramalikazi/spring-async-cache
Try this out, improve end-user response time and shine... :) Enjoy...