Event Processing

Event Stream Processing or Complex Event Processing?


Event Stream Processing and Complex Event processing is computing on live event streams. Event streams can contain a wide variety of time based occurrences (events). These can range from work orders to alarms, warnings or events associated to process conditions, items in your workflows, events from your assets et cetera et cetera.

Very often event streams contain valuable information when events are occurring in a certain sequence of time or in a specific frequency. In certain cases it is also valuable to know when event streams do not follow a known pattern.

An example of an event pattern for a windmill is:

  • An increase in wind speed;
  • An stable rotor speed; and
  • An increase in bearing temperature.

This may be indicative of a bearing failure. The example is very clear because it correlates only a few events in the time domain and none of the other data that is also contained within the data/event stream of the windmill is shown.

So why both Event Stream Processing and Complex Event processing?

The definitions for Event Stream Processing (ESP) and Complex Event Processing (CEP) are very often used to describe the same. There is however a subtle but important difference:

  • CEP is the process that combines events from multiple sources to infer new events that describe patterns or other complicated causal relationships (circumstances).
  • ESP is a technology aggregation term that includes event visualization, storage (databases) and event processing language(s) (=CEP).

Event processing languages – CEP – are key in event Stream Processing and one of the enablers for Operational Intelligence applications. These are applications that tell you what is happing now and will happen, based on streaming data.  This is very different from traditional BI and BI tools that inform you about things that have happened in the past.

Complex in Complex Event Processing does not mean complicated!

We have been building on our CEP engine since 2003 and have been able to simplify the user interface for domain (application) centric usage. A highly intuitive interface is available for defining patterns of interest on event and alarm streams:



No .. our CEP engine comes with plenty ESP features. Beside visualization of event and alarm streams (and important the aggregation of those streams in real-time) we also provide capabilities to link this to a wide variety of data sources – the interfaces to Microsoft Azure Event Hub .. streaming millions of events per second … the latest.

More details in next BLOGs.