Sergej Khackimullin - Fotolia
5 ways to get complex event processing right
How should software teams make the best use of complex event processing, and how can they implement it successfully? This article looks at five fundamental truths about CEP.
As an organization's software architecture becomes increasingly complex and distributed, and devices generate data on all manner of events, it's difficult to keep everything on track without capacity problems, compatibility roadblocks and piles of unused data. Complex event processing deals with this problem.
CEP can aggregate data in intelligent ways, so organizations then analyze and derive actual value from disparate streams of information and ensure software functions on any number of devices. There are five fundamental things to bear in mind when it comes to CEP, from the best time to use it to how the CEP engine will fit with the organization's needs from its data in the future.
1. Real-time processing is better than post-event CEP
CEP can be used to analyze data after an event. For example, it's become a tool to execute forensic investigations of computer attacks and failures. However, the real value lies in its ability to carry out real-time event processing.
Organizations investing in complex event processing should make real-time analytics the goal. After all, it is far better to prevent a hack through predictive data than to try to figure out what happened after the damage is done.
2. Edge processing contains the capacity sprawl for low-level IoT
IoT devices generate huge amounts of data, but not all data is equally useful. An IoT sensor that reports temperature every five minutes can be safely ignored if all it reports is that the temperature stays within prescribed limits, for example. If the device itself has little intelligence, use edge processing to dump the unneeded data as soon as possible. This event-processing step close to the data source minimizes the flows of information onto and across the network and improves analytics response times.
3. Standards are there for a reason
Data comes from multiple sources in many event-based applications, and each source sends data in its own format. Often, the data format from a source won't adhere to your basic standards.
Rely on edge processing to trap data and convert it to a standardized format, such as an open database format or XML. Organizations can then derive real value from that data, as the standardized format makes it possible to analyze information regardless of its source.
4. Work with CEP vendors who have the future in mind
The majority of CEP systems and platforms are rules-based, which means the CEP engine mainly depends on if-this-then-that logic. Such rigid constructs will become a constraint to how well CEP can operate. The ability of a CEP platform to adapt and learn, through artificial intelligence, while an organization changes over time will be critical for CEP's value. An adaptable CEP platform can be used to identify new threats and opportunities, as well as deal with them. As part of due diligence during CEP selection, make sure that the CEP vendors you consider have a roadmap that moves beyond rules-based logic.
5. Minimize mistakes with upfront effort
CEP is already becoming a must-have in the application architecture and IT management toolbox. However, it is easy to choose the wrong system or implement the right system in the wrong manner. Planning is key, along with the correct choice of CEP engine.
Think about what the company will need in the future from CEP, and whether the CEP engine can meet those needs. Replacing the CEP engine, or making extensive modifications with heavy reliance on plugins, adds up to a lot of work that was avoidable during the selection process.