Process API Architecture Overview

PAPI is a Java API that allows you to build a client to connect to the engine and perform operations on the deployed processes.

When PAPI is initialized, the connected user is authenticated against the data in the directory service. Once authenticated, the Java client using PAPI can interact with any of the engines configured in the directory service.

PAPI must connect to a Process Execution Engine only to search for or operate on process instances and deployed processes. It does not need to connect to an Engine if a request or operation requires only data stored in the Directory Service.

The following operations do not require a connection to the engine:

List available views
Search for a specific view
List available presentations
List participants in the organization

You can successfully execute these operations even when the Process Engine is down.

PAPI can connect to one or more engines at a time, provided they are configured in the same directory service. When a client makes a request, PAPI automatically routes this request to the corresponding engine.

The following diagram shows interaction between a PAPI client, the Directory Service and the Process Engine in runtime. The diagram shows a custom Java client and Oracle BPM Workspace that is also a PAPI client.

Figure 1. PAPI Components Runtime Diagram