The BPMN2 Engine will have a clear focus on being light weight and easy to use for Java developers..

Features
This is the heart of the BPMN2 project. It's a Java process engine that runs BPMN 2 processes natively. It will have the following key properties:

1. Allows user updates to be combined with process updates in a single transaction
2. Runs on any Java environment like Spring, JTA, standalone with any form of transaction demarcation.
   
3. Easy to get up and running with the setup utility
4. Built to support the cloud scalability from the ground up
5. Very simple to add new custom activity types and complete dedicated process languages
6. Rock solid
7. Extremely fast
8. Transactional timers
9. Asynchronous continuations
10. Hidden event listeners for decoupling software technical details from business level diagram
11. Ability to test process executions in isolation in a plain unit test
    A couple of features deserve extra highlighting because they improve the collaboration between non technical business users and technical developers.

First of all event listeners. That features allows to execute an out-of-the-box action, a piece of custom Java code or a script upon certain events in the diagram. This means that developers can decorate a process with extra technical details that don't show up in the diagram. This improves the collaboration between business people and technical people as the business people will not be confronted with technical aspects in the diagram.

The Process Virtual Machine
The Process Virtual Machine is the architectural base layer of the BPMN2 Engine. It allows for easy pluggability of activity types, features and complete process languages. The core BPMN2 team is dedicated to building out the native BPMN 2.0 process language. BPMN 2.0 will be the predominant language used on the business side of BPM. And hence we want to simplify the bridge between business modeling and making those processes executable.

But there are many developer oriented process languages that are very easy to build on the Process Virtual Machine. There are numerous other languages that could be build on top of the Process Virtual Machine. We even envision companies building their own variants of domain specific process languages.

The core BPMN2 engine of course includes

• persistence (JPA-based with pluggable variable persistence)
• transaction support
• auditing
• history log
• basic process instance migration
• etc.

Also support for domain-specific nodes and powerful rules integration.

Human tasks

Independent human task service for managing human tasks based on the WS-HumanTask specification.
Simple web-based human task web console supporting the task life cycle (claim, start, complete, etc.) and custom task forms.

 Eclipse-based tooling for creating BPMN2 processes

Eclipse-based plugin that allows developers to graphically create BPMN2 processes, including basic validation, testing and debugging.

Web-based tooling for creating BPMN2 processes

We'd like to continue the integration of the web-based BPMN2 editor based on the open-source Oryx editor.

Process repository

Knowledge repository for storing process definitions.
Simple web-based repository management console for storing process definitions, versioning, releasing, etc.

Process management console
Simple web-based console for starting processes, managing running instances, checking current state of one specific instance, aborting, etc.

Reporting
Customizable reports using BIRT
Simple web-based console for viewing real-time reports

Installation script (and demo setup)
Documentation