Real-time vs Near RT
Why both are viable options and have their own purpose.
In the world of data processing, there are more ways to do it than ever. The purpose for your data processing will determine the solution you need. There are two basic forms of data integration, real-time and near real-time (NRT). In many cases, you're processing old or archived data and time isn't of huge importance. However sometimes processing tasks are crucial, and the answers must be delivered in seconds to maintain their value. There are differences among real-time and near real-time that will inform your decision of which is the best option for you.
Real-time data integration is just what it sounds like. The program collects and aggregates the data as soon as changes are made in the database. Real time processing requires a continual input, constant processing, and steady output of data. A great example of real-time processing is data streaming, radar systems, customer service systems, and bank ATMs, where immediate processing is crucial to make the system work properly. 
Near Real-time (NRT) is data integration that occurs periodically. Data is collected immediately, but it isn't necessarily integrated instantly. NRT data integration solutions can be programmed to process the data it collects every day, every hour, or even every few minutes, at the user's discretion. Near real-time processing is when speed is important, but processing time in minutes is acceptable in lieu of seconds.
In choosing between the two, variables like system requirements and system setup come into play. Real-time processing is meant for large volumes of data in constant motion, requiring a high-functioning system. If you need real-time processing but your system isn't quite up to the task, an upgrade may be needed before real-time data integration can happen. However, don't let the instantaneous nature of real-time, near real-time solutions is no slouch in the processing speed. Depending on your source systems near real-time can reach real-time speeds.