Continuing to look at the Introduction to Oracle Event Processing book, by chapter 6 the books has covered the key principles and ideas for building a CEP solution, and we’re now need to consider deployment. But refreshingly the book also takes on a number of non functional requirement (NFR) areas such as the issue of monitoring, a subject area that many technical books tend to ignore. The attention to monitoring is admittedly driven by the fact that Event Processing is inherently sensitive to timing and system loading etc and will obviously have a direct impact on what outcomes are produced.
As the book takes you through aspects of building a simple solution – the CEP equivalent to writing ‘Hello World’ it would be great if the authors could make the implementation available for download, so you could go straight into deployment.
Chapter 7, then takes us into other performance improving aspects such as how to get event enrichment data, and importantly exploit caching to drive the performance. If you’re familiar with Coherence then this aspect should be pretty easy to get to grips with, and the book actually focuses on the OEP aspects of the setup. If you don’t know Coherence, you’d do well to look at additional sources of information.
Chapter 9 is a natural evolution of 8 as further develops performance thinking with clustering and with it High Availability dimensions.
Before looking at High Availability (HA) and scaling the book drives back into more advanced scenarios with CQL by introducing Java into the syntax.