Oracle BPM Suite 11g: Advanced BPMN Topics: This tutorial reaches the parts that standard manuals don't, taking you deep into advanced BPMN topics for Oracle BPM Suite. With a practical approach and logical explanations, it will make you a maestro of BPMN.

In this practice activity, we will build two processes and call one from the other using correlation . The example that we will use is illustrated in the following diagram. We will create a calling process that will send a number to the called process . The called process will square the number (multiply it by itself) and return the result to the calling process.

We will run multiple instances of the calling process, which will start multiple instances of the called process—one each. We will observe that the correct result is returned to each instance of the calling process.

In this practice activity, we will extend the previous activity to include correlation when we call the Called process multiple times inside a loop. This will allow us to explore a slightly different area of correlation—scoped conversation s and correlation when the process is participating in multiple conversations. In this activity, our Calling process starts a number of sub-processes . Each of these in turn starts an instance of the called process, as illustrated in the following diagram:

Let's start by creating some additional data types we will need now:

In this practice activity, we will explore how to use signals to create publish/subscribe style communication between process instances, as illustrated in the following diagram. You may recall that when using signals, the sending process does not know which (if any) other processes will receive the signal.

In this practice activity, we want to explore how to use reusable sub-processes. You may recall from Chapter 1, Inter-process Communication, that a reusable sub-process follows the pattern of starting with a Start node with type None and ending with an End node with type None. They are called from a parent process with the Call activity.

The Called process that we created in the first two practice activities actually follows this pattern; however, we did not use a Call activity to invoke it in those practice activities.

You may also recall from Chapter 1, Inter-process Communication, that a reusable sub-process does not have access to its parent's variables unless they are passed in by the parent as arguments . In a reusable sub-process, you define the input and output arguments for the process at the process level, not on the Start and End nodes.

The Receive and Send tasks in the Called process we saw earlier are not the mechanism used for passing data in and...

In this chapter, we have practiced the skills learned in Chapter 1, Inter-process Communication. Specifically, we have looked at how to use correlation with the default conversation to communicate between instances of processes. We have seen that each of the instances of the calling process started a conversation with an instance of the called process, and each reply was sent back to the correct caller.

We explored the use of a scoped conversation and how correlation works with an advanced conversation (not the default conversation). We saw how to define the scoped conversation inside an embedded sub-process.

We also saw how to create and use a reusable sub-process and reviewed some important characteristics and limitations of reusable sub-processes.

The embedded sub-process is a very useful construct for dealing with collections of data. In the next chapter, we will move on to explore how to work with arrays and collections of data in more detail.