Looking at Bill Gassman's presentation in the Gartner EPS on BAM - he talks about the BAM market and partition it to:
- 65% - embedded inside vertical solutions
- 20% - embedded inside BPM (+ ESB) middleware/products
- 10% - embedded inside BI products
- 3% - embedded inside IT operations (e.g. BSM)
- 2% - general purpose BAM products
Gassman's interpretation of BAM is quite wide (I have somewhat narrower interpretation), and covers most of the EP types of applications, so let's take it as a starting point. Without commitment to the exact numbers, the order is consistent with my observation on this market. While the early adopters used stand-along engine, and built applications on top of it, this will become a relatively small route to market, and the segment that will grow most is that of EP technology embedded inside vertical solutions, we see signs of this is multiple industries now. The second largest segment is EP technology embedded inside middleware, we see that the big players in this area are taking this approach, the rational behind it is twofold - from the middleware point of view - EP capabilities is now becoming a must, due to competitive pressures, and from the ROI to customers POV - EP applications are typically not isolated, and the biggest investment is to connect them with the consumers/producers of events, thus application integration middlewares with adapters to multiple systems may assist. There will be always a market for a stand-alone event processing technologies, and this market can be segmented to "general purpose" engines, and optimized for special purpose ones..... I am not sure that what I am writing now reflects the current reality, but it certainly reflect the trend....
More on the role of general and specific frameworks - later.