15 Threading and Clustering Topics

This chapter describes how to use threads in WebLogic Server as well as how to program applications for use in WebLogic Server clusters.

This chapter includes the following sections:

Using Threads in WebLogic Server

WebLogic Server is a sophisticated, multi-threaded application server and it carefully manages resource allocation, concurrency, and thread synchronization for the modules it hosts. To obtain the greatest advantage from WebLogic Server's architecture, construct your application modules created according to the standard Java EE APIs.

In most cases, avoid application designs that require creating new threads in server-side modules:

  • Applications that create their own threads do not scale well. Threads in the JVM are a limited resource that must be allocated thoughtfully. Your applications may break or cause WebLogic Server to thrash when the server load increases. Problems such as deadlocks and thread starvation may not appear until the application is under a heavy load.

  • Multithreaded modules are complex and difficult to debug. Interactions between application-generated threads and WebLogic Server threads are especially difficult to anticipate and analyze.

In some situations, creating threads may be appropriate, in spite of these warnings. For example, an application that searches several repositories and returns a combined result set can return results sooner if the searches are done asynchronously using a new thread for each repository instead of synchronously using the main client thread.

If you must use threads in your application code, create a pool of threads so that you can control the number of threads your application creates. Like a JDBC connection pool, you allocate a given number of threads to a pool, and then obtain an available thread from the pool for your runnable class. If all threads in the pool are in use, wait until one is returned. A thread pool helps avoid performance issues and allows you to optimize the allocation of threads between WebLogic Server execution threads and your application.

Be sure you understand where your threads can deadlock and handle the deadlocks when they occur. Review your design carefully to ensure that your threads do not compromise the security system.

To avoid undesirable interactions with WebLogic Server threads, do not let your threads call into WebLogic Server modules. For example, do not use enterprise beans or servlets from threads that you create. Application threads are best used for independent, isolated tasks, such as conversing with an external service with a TCP/IP connection or, with proper locking, reading or writing to files. A short-lived thread that accomplishes a single purpose and ends (or returns to the thread pool) is less likely to interfere with other threads.

Avoid creating daemon threads in modules that are packaged in applications deployed on WebLogic Server. When you create a daemon thread in an application module such as a servlet, you will not be able to redeploy the application because the daemon thread created in the original deployment will remain running.

Be sure to test multithreaded code under increasingly heavy loads, adding clients even to the point of failure. Observe the application performance and WebLogic Server behavior and then add checks to prevent failures from occurring in production.

Using the Work Manager API for Lower-Level Threading

The Work Manager provides a simple API for concurrent execution of work items. This enables Java EE-based applications (including servlets and EJBs) to schedule work items for concurrent execution, which will provide greater throughput and increased response time. After an application submits work items to a Work Manager for concurrent execution, the application can gather the results. The Work Manager provides common "join" operations, such as waiting for any or all work items to complete. The Work Manager for Application Servers specification provides an application-server-supported alternative to using lower-level threading APIs, which are inappropriate for use in managed environments such as servlets and EJBs, as well as being too difficult to use for most applications.

For more information, see "Using Work Managers to Optimize Scheduled Work".

Programming Applications for WebLogic Server Clusters

JSPs and servlets that will be deployed to a WebLogic Server cluster must observe certain requirements for preserving session data. See "Requirements for HTTP Session State Replication" in Administering Clusters for Oracle WebLogic Server for more information.

EJBs deployed in a WebLogic Server cluster have certain restrictions based on EJB type. See "Understanding WebLogic Enterprise JavaBeans" in Developing Enterprise JavaBeans, Version 2.1, for Oracle WebLogic Serverfor information about the capabilities of different EJB types in a cluster. EJBs can be deployed to a cluster by setting clustering properties in the EJB deployment descriptor.

If you are developing either EJBs or custom RMI objects for deployment in a cluster, also refer to "Using WebLogic JNDI in a Clustered Environment" in Developing JNDI Applications for Oracle WebLogic Server to understand the implications of binding clustered objects in the JNDI tree.