Book Image

Mastering JBoss Drools 6

By : Mariano De Maio, Mauricio Salatino, Esteban Aliverti
Book Image

Mastering JBoss Drools 6

By: Mariano De Maio, Mauricio Salatino, Esteban Aliverti

Overview of this book

Mastering JBoss Drools 6 will provide you with the knowledge to develop applications involving complex scenarios. You will learn how to use KIE modules to create and execute Business Rules, and how the PHREAK algorithm internally works to drive the Rule Engine decisions. This book will also cover the relationship between Drools and jBPM, which allows you to enrich your applications by using Business Processes. You will be briefly introduced to the concept of complex event processing (Drools CEP) where you will learn how to aggregate and correlate your data based on temporal conditions. You will also learn how to define rules using domain-specific languages, such as spreadsheets, database entries, PMML, and more. Towards the end, this book will take you through the integration of Drools with the Spring and Camel frameworks for more complex applications.
Table of Contents (18 chapters)
Mastering JBoss Drools 6
Credits
About the Authors
About the Reviewer
www.PacktPub.com
Preface
Index

When not to use a rule engine


Every project, in one way or another, can benefit from using business rules. They are highly performing, easy to change, and self-explanatory software components. However, there are a group of conditions that a project might have that would make use of business rules a bit of overkill. Some of the characteristics that make a project benefit the least from business rules are shown in the following:

  • There are very few, self-contained rules involved in the project: If the business rules identified in the requirement gathering are very simple and span about one or two objects at most, we don't need a rule engine to run them. A good rule of thumb is that if we can write the business rules that we need as the pseudo code in less than a page and with less than two nested if-then clauses, we might not need a rule engine at this particular time.

  • The business logic doesn't change often: If changing rules at runtime is not going to be needed but the logic is still complex, a rule engine might still be a good idea. However, if the complexity behind the rules is not that high and we can assume it will remain that way for a long time, we might not need a rule engine.

  • A very strict control of the execution flow is crucial for the application: As we stated before, a sequence-flow control is not provided when we execute our business rules. If the business logic behind the business rules depends a lot on a strict set of steps that need to be executed sequentially, business rules might not be the right fit. However, if it does change frequently, perhaps a business process would be worth considering.

It is still a responsibility of the project team to determine whether business rules might be a good fit even if these conditions are met. After all, our experience can lead us to think that the amount of rules has a big chance of growing in the future or there might be situations where the rules will eventually need to change more frequently. Each project has its own unique characteristics and it might be that a project with no need for business rules right now cannot be thought without them in the future.