Event Processing

Event processing is the generic term that subsumes both event stream processing (ESP) and complex event processing (CEP). Event processing engines are used to support operational business intelligence and automated and semi-automated decision making where latency needs to be minimised; in other words, where there are genuinely "real-time" needs. Such requirements occur in capital markets, fraud detection and prevention, network management of various types from railways to pipelines, business activity monitoring, telecommunications tariff look-up, point-of-sale optimisation and many others.

Although the initial development of event processing engines started as long ago as the late ‘90s, and the first products emerged early in this century, it is only now that event processing is really starting to take off. However, in so far as vendors and products are concerned, we find there are:

• a variety of different approaches that are used
• no consistency as to the languages that may be deployed for development purposes
• wild and misleading claims made about performance
• differences between products aimed at business analysts versus developers
• different technologies used for event storage
• issues over the consistency of results
  • to name but a few of the unresolved issues in event processing.

Worse, there are general-purpose products and niche products, and products that only do event stream processing or complex event processing. There are also vendors that only focus on certain market areas who might expand into other areas in due course.

This Bloor Research report describes the technology and what it is good for, including providing a series of use cases. In other words: what event processing is, how it works and where you might use it. It also describes and resolves (in so far as that is possible) the issues mentioned. In addition, the report compares and contrasts the various vendors/products in the market, including: Agent Logic, AleriLabs, AptSoft, Coral8, GemStone, Leanway, Kaskad, Progress, SeeWhy, Streambase, Syndera, Tibco and the open source Esper. It also considers specialised solutions from the likes of Microsoft, SAS and Sybase as these vendors seek to move into this marketplace.

If you have read this paper and have an opinion on it's content, please login and post a review for others to see.