The evolution to the event-driven architecture

Computerization of our economy has played a critical role in the development of our society. The development is not limited to business models; it has elevated the expectations of individuals and groups. The latter has only recently become apparent from the rise of the models of social networking and the technologies such as the smart phone. It can be difficult to predict, much less understand, the characteristics of the next movements in technology, corporate behavior and the actors that will participate. One thing remains clear. As the technology progress, economic ecology (businesses and government entities) progresses onward. For instance, most business historians and economists consider the rise of Wal-Mart to be a phenomenon. Wall-Mart owes a large portion of its success to technology. It is well known that their technology contributed to a unique shift in models for retail and supply chain management.

Business models and architectures have not remained stagnant. Most recently, we see a shift in advanced segments of the world:

  • Consumers are expecting highly personalized and time-relevant responses from retail and service providers
  • Businesses need more visibility into the conditions and events that shapes their operations, especially in the areas of risk

Clearly, companies that provide this complex personalized and time and location critical consumer responsiveness will be the leaders. Moreover, companies that operate at these advanced corners of technology are the most competitive and profitable. They attract the brightest employees and executives and have the best business opportunities.

New architectural requirements

For over 20 years, technical architecture has undergone cyclic shifts, arriving at every 4 to 8 years. Shifts have evolved from mainframe to client server, to multi-tiered, to the internet and onward. It is apparent that another shift is taking place. The shift is the addition of Event-driven approaches to Business Process Management. For computing architecture, the shift is from service oriented architectures (SOA) to Event Driven Architectures (EDA). It is obvious that EDA will be critical to meet the newly emerging needs we have alluded to.

Innovations’ implementation of the EDA is known as the Visual Rules Enterprise Suite. This framework and its components provide the basic functionality organizations will need to implement an event-driven process. The event-driven process is one of the most important process usage patterns. Organizations use the pattern to interact and respond to a growing volume of business events and transactions in an agile, proactive way. Event analysis is an emerging area of business rules and process modeling that develops support for the decision-based processing of enterprise-significant events. Because events can be produced by devices, it is also increasingly an important part of strategies for the evolving ‘internet of things’.

So, business processes must accommodate event-driven, actionable situations. These include key events or combinations of events. The process model should be constructed to react with the correct sequence of activities. As processes become increasingly discovered, mapped, managed, and digitized there is an increasing need for external event processing. A complete discussion of the Event Driven Architecture is beyond the scope of this article; however we will describe the relationship between EDA and the business rules that implement event processing for improved decision making and governance enforcement.

Event processing on the Visual Rules Enterprise Suite

Chandy and Schulte [1] define “A business event is an event that is meaningful for conducting commercial, industrial, and governmental or trade activities.” An event is Boolean, in nature, it either happened (True) or not at all (False). The event is meaningful because it might affect a business process, as an external message or channel, which one or more processes must consume, activate and respond to. The event-driven process is activated or responds to the event processing facility.

As shown in figure 1, Business Event Processing in combination with a Business Rules Management system can create an effective combination of timely event pattern detection and dynamic business process execution. Business Event Processing simplifies the construction of the event-driven business process usage pattern. Event processing notifies channels with actionable business situation. Typically they signal the business process to respond. Through its core technology, Visual Rules, the Innovations framework supports graphical, nonprocedural user interfaces that business analysts and managers use to define the event processing interactions and actions.

Figure 1, In an Event Processing Systems, the cloud of events is processed by business rules.

Figure 1, In an Event Processing Systems, the cloud of events is processed by business rules.

David Luckham [2] defined the event space as a cloud because events are not nicely organized. “They do not necessarily arrive at the enterprise in the order they were created or in their causal order… they form an unorganized cloud of events”. Beyond sending a message to a process, Figure 1 raises a number of important issues. How will the event processing capability know what events to send to the processes? Can a process contribute events to the cloud? Event processing systems generally implement the three key concepts:

  1. Event detecting functions or sensors of input events,
  2. Event Decision Logic and
  3. Event channels for consuming processes.

The goal of the Event Process is to implement event processing systems and control event-driven business processes. It also provides a common framework for specifying event processing solutions.

Event Decision Logic

In the Event Management component of Innovations Framework, there are four classes of decision logic for event processing:

  1. Filtering,
  2. Processing
  3. Categorizing
  4. Routing

In the ‘real world’, event producers may not follow expected patterns. Events may differ in the expected syntax (structure), semantic meaning, or both. In a business process, a single starting or intermediate event might be part of complex composition of events happening at different times. Event decision logic identifies an event from patterns within the cloud of events. With the logic of the rules, it also derives new events. The decision logic produces these derived events and decides where and how these events should be made available (impact management). Finally, rules encode the intelligence that decide, based on infrastructure parameters, how to escalate or further route the event.

Filtering decisions select the event. Conditions for a relevant business event can be dynamic. Events of interest might be limited to a time range. For instance, we might only select events arising during week days after a holiday. Other dynamic filtering conditions might include: declining market conditions or weather patterns. The point is that what is an important event during one condition might not be important during another. Filtering decision logic decides this.

Event Processing Decision logic transforms, validates, or enriches the event attributes. An incoming event might be the transmission of an X11 document by a trading partner. The decision logic will probably use a validation step to verify the content of the document, a transformation step to place it in a form that the business can accept and an enrichment step to simplify processing of the document.

Event categorizing decision logic connects the instance of an event to other events. An event can be dynamically connected to others. For instance you can use this logic to associate events with others.

Event routing logic identifies the channel for the event. Channels are nodes that connect event producers to the business process. An event can be conditionally routed to multiple processes. On the process side, in BPMN or an enterprise service bus, the channel might be denoted as a start or intermediate message.

At Innovations, we are using the Visual Rules Enterprise Suite in a pilot project to manage customer electric vehicle’s charging needs. These comprehensive, event-driven systems, often called E-Mobility, must integrate many event-driven and dynamic factors. One event-based, system requirement is that conditions of the charging stations are sent to the central server. The platform uses rules to route these events to a central server. A comprehensive E-Mobilitiy platform must accommodate various events with various destinations. Because business analysts have visibility and access to the Rules, Managing E-Mobility with the Visual Rules Enterprise suite is remarkably straightforward.

Event channel

An event channel publishes, or feeds, events and event streams to business processes. Once the event has been directed to a channel event processing is complete. The event channel might transmit multiple events from different event processes. Alternatively, it might transfer events combined from several processes. Event Decision Logic orders the events from processes and combines the events into a set.

Event repository

Figure 2: an event repository is a critical part of event processing.

Figure 2: an event repository is a critical part of event processing.

Another responsibility of an event processing environment is flowing through the history of the events for retrospective event processing. There can be decision logic associated with event retention policies that determine the duration and filtering conditions for retained events. A repository permits the discovery of historical event patterns.

Summary

With the Innovations´ Framework and the expressive power of Visual Rules Enterprise Suite, analysts, business and technical, can become more self-sufficient and empowered to create the new event driven solutions to technical and business problems, with less involvement from IT. In many of today’s organizations, the business analyst, manager or subject matter expert creates and maintains processes and applications and they do this with skills previously considered technical. These organizations operate in areas where multi-week development cycles would be intolerable. These areas include risk management, fraud detection and complex areas of finance and investment- areas that are addressed by the event processing metaphor. They make (again, would have been considered) highly technical changes to the applications. This is especially true in the areas of the Event Driven Business Process.

 

[1] K. Chandy, W. Schulte, Event Processing: Designing IT Systems for Agile Companies, McGraw-Hill Osborne, 2009
[2] David Luckham, The Power of Events: An Introduction to Complex Event Processing in Distributed Enterprise Systems, Addison-Wesley Professional, 2002

About The Author

Tom Debevoise

Tom Debevoise was a regular contributor to our blog until 2013. His interest is in building practices that empower the business analyst or expert to use a framework of software and methods to create their own process, services and managed events. Tom is the the author of several books, "The Data Warehouse Method," "Business Process Management with a Business Rules Approach" (2006), and "A Microguide to Process Modeling in BPMN".

Leave a Reply

Your email address will not be published.