• Business Technology Consulting Graduate Blog
• Consulting Blog

Complex Event Processing and Enterprise

What is complex event processing? For the uninitiated let me start by scratching the surface of some theoretical computer science as an introduction. The solvability of computationally complex programmable algorithms in software design is governed by the ability to benchmark them against non-deterministic polynomial (NP) time, polynomial (P) time, or the grey fuzzy cloud of the unknown complexity known as NP-Complete. This school of thought, however, is well known and there’s even a prize of one-million-dollars for the person who can derive a proof providing evidence to show that solvable problems in complexity class P are equal to those in class NP – i.e. P = NP. But what has this got to do with a corporate enterprise? And what has computational complexity got to do with complex event processing?

Let’s consider the current state of the economic downturn as an example. The repercussions of the subprime mortgage crisis across the Atlantic come through in short bursts, we’d typically experience ripples which were for the most part predictable – such as those coming through the conservatorship of the Federal National Mortgage Association (you’ll probably know of them as ‘Fannie Mae’) and Federal Home Loan Mortgage Corporation (again, commonly known as ‘Freddie Mac’). But as the big corporate names started to fall – such as AIG (acquired by US Federal Government), Lehman Brothers (bankruptcy), and Merrill Lynch (acquired by Bank of America) – the ripples were coming through stronger and at an accelerated pace. Now as we are in the midst of a global recession, the volatility and shear unpredictability of the economy means for business to stay afloat they must remain competitive in preparation for the upturn. In order to do so, enterprises must operate with an agility to respond proactively to oscillations in the economy, helping them to address the challenges in a rapidly evolving economy.

Fundamentally, tackling unpredictable events – such as rapid changes during a recession – is about quantifying the impact of such an event and thus being able to derive a weighted approaching to prioritising the processing of events in the least disruptive manner. A common approach to this is to conceptualise an event, more specifically to recognise the hallmarks of an event and to better understand what the repercussions could be. This forms the basis of complex event processing. More formally complex event processing can be explained as “a technique that helps discover complex, inferred events by analysing and correlating other events”.

So how does complex event processing relate to IT within the corporate enterprise? With corporate software architecture developing into a service-oriented architecture, and that coupled with the rapid adoption of radio frequency identification, event processing is well suited to fit in enterprise information systems in terms of facilitation of event aggregation into high-level actionable information, and improving the responsiveness. And as you can see, complexity in IT isn’t just about overcoming the challenges of building complex software systems by designing efficient algorithms – it’s also about understanding what happens when minor ‘events’ come together to cause a major complex event, and how best to tackle these in a methodical order.

I’d be fascinated to hear your thoughts on this, and to hear of any examples of complex event processing occurring at your workplace.

Comment Moderation is enabled. Your comment will not appear until approved.

Post Your Comments

1. Name:
2. Email Address:
3. Website:
4. Comments:
5. Enter text that you see in the image:
   
6. Remember My Information:
7. Subscribe:

If you subscribe, any new posts to this thread will be sent to your email address.

Complex event processing sounds interesting. Another interesting point you touched on is the "unpredictable events". You may find Nassim Taleb's the black swan book interesting. It suggests economic boom/crashes although not predictable will always happen. He goes further in his books to suggest techniques for playing the market to profit from this fact.

http://www.amazon.com/Black-Swan-Impact-Highly-Imp...

sidenote: I don't like the authors writing style, he comes across as very arrogant but the book gets you thinking about some interesting concepts though I question some of its conclusions. # Posted By Ryan Hamilton | 6/4/09 7:20 PM

If you're interested in reading more about CEP, then I'd like to introduce you to the work of long-time CEP guru (and Stanford University professor) David Luckham. His work and industry knowledge is a great inspiration to me, and it actively encourages me to continually look at my work holistically – i.e. looking at it in parts rather than as a whole - and motivates to think “innovation” whenever I tackle a piece of work. # Posted By Sham Mitra | 6/13/09 10:31 PM

On the back of this blog post I came across an interesting article which I felt compelled to share with the readers of this blog. The article tries to shed light on the solvability of P=NP by using the Riemann Hypothesis as an example. # Posted By Sham Mitra | 7/3/09 6:12 PM