After some interesting discussions in the blogosphere, the results of our less-than-scientific Poll: Where is CEP Relative to the Gartner Hype Cycle? are in!  Two-thirds of our passionate-about-CEP readers believe that CEP still is in the very early stages of maturity.

Only 6% of the those who responded to the poll, conducted from July 3 thru July 12th, voted that CEP was mature.   If you include those who consider CEP getting close to maturity, 18% of our readers who voted said that CEP was in the final stages of maturity.

How is it possible that 31% of the folks who responded believe that CEP is in the Gartner-defined Technology Trigger stage of maturity, while 6% believe CEP is at the other end of spectrum, in the Plateau of Maturity Phase?

During the poll I received a question from a colleague who asked me if I “still loved CEP?” and “why are you trashing the entire industry that you love?”

Frankly speaking, I have enjoyed a passion about event processing since my early days at Sprint, circa 1993, during the NSFNET transition to the commercial Internet.   Then, as today, we hoped for the same goals and objectives that network and security management people seek to achieve; high confidence in actionable alerts with a very low false alarm rate, all based on processing myriad distributed networking events, sometimes referred to today as sense-and-respond networking.

Today, we are good at “sensing”.  Events are created, perhaps trillions upon trillions a second globally.   No one knows the exact number of events the world’s networks generate in a single second, much less in a day or a year.      Yet, we are quite good at producing events.

What we do know is that we do not yet have the technology to listen to myriad events and determine complex events and situations with high confidence.   At best, we can detect, sense-and-respond, to simple events and primitive situations.

On the other end of the maturity curve, there have been some advances.  Some of the notable progress has been in the event stream processing (ESP) space.    ESP is an importart part of the equation but it is nowhere close to the entire solution because rule-based stream processing is at a very low level in most sense-and-respond decision-making models.  Higher level inference requires more sophistication.

Two-thirds of our readers believe that CEP is still in the very early stages.  The majority of our readers envision CEP as a technology, or set of technologies, to solve myriad complex event processing problems and they know we have a long way to go.     On the other hand and with just as much passion, about one-in-six readers think that the technology is mature, and we are at the end of the CEP maturity cycle.

My crystal ball is just as foggy as yours on the future of CEP – but here on The Complex Event Processing Blog, we continue to work hard to “keep it real” for our readers.


  1. please take a look at the article
    on event driven process and performance management, which i had written in late 2002.

    the reason i think is that CEP has been around for some time (and hence may appear mature purely from time standpoint) but may also be considered early from an adoption standpoint, is that business solutions are lagging behind the technology. CEP needs to be de-mystified for business users – that i think is the only way to accelerate it through the technology curve.

Comments are closed.