Book Image

Drools JBoss Rules 5.X Developer's Guide

By : Michal Bali
Book Image

Drools JBoss Rules 5.X Developer's Guide

By: Michal Bali

Overview of this book

<p>Writing business rules has always been a challenging task. Business rules tend to change often leading to a maintenance nightmare. This book shows you various ways to code your business rules using Drools, the open source Business Rules Management System.<br /><br />Drools JBoss Rules 5.X Developer's Guide shows various features of the Drools platform by walking the reader through several real-world examples. Each chapter elaborates on different aspects of the Drools platform. The reader will also learn about the inner workings of Drools and its implementation of the Rete algorithm.<br /><br />The book starts with explaining rule basics, then builds on this information by going through various areas like human readable rules, rules for validation, and stateful rules, using examples from the banking domain. A loan approval process example shows the use of the jBPM module. Parts of a banking fraud detection system are implemented with the Drools Fusion module which is the complex event processing part of Drools. Finally, more technical details are shown detailing the inner workings of Drools, the implementation of the ReteOO algorithm, indexing, node sharing, and partitioning.</p>
Table of Contents (22 chapters)
Drools JBoss Rules 5.X Developer's Guide
Credits
About the Author
About the Reviewers
www.PacktPub.com
Preface
Setting Up the Development Environment
Creating Custom Operators
Dependencies of Sample Application
Index

The KnowledgeBase partitioning


Drools supports parallel execution mode. One session can be executed by multiple threads.

Note

Note that this feature is not supported in Drools 5.5.0.Final. It might be reintroduced in future; for now, keep this in mind and take this information as a curiosity.

The Rete network is split into multiple partitions. Each partition is handled by a PartitionTaskManager object. It manages a list of suspended propagations and makes sure that only one of them is being executed at a time over this partition. When a fact is propagated through the network, it may go through one or more partitions. Once a propagation reaches the boundary between two partitions, the other partition's PartitionTaskManager is notified and the current propagation is transferred to its list of propagations. The suspended propagation then waits in this list until the other partition manager is ready to take it further.

Note

Each knowledge session has a unique set of its own PartitionTaskManager.

How...