As we have seen in the previous section, implementing edge analytics into our IoT design has advantages that outweigh the costs and complexity. However, before we can build an edge analytics system, we need to know what the key benefits are.

The following is a list of those benefits (note that these are the same key benefits as described in the Edge computing section):

• Privacy
• Latency
• Reliability

Let's take a look at an example. In the following diagram, we have designed an advanced security application using an edge analytics architecture. As you can see, we have a Raspberry Pi connected to a webcam, automated door, and a light sensor:

Our advanced security application uses the webcam and Raspberry Pi to recognize a visitor trying to gain access to the building. Based on the match of our visitor with a repository of known visitors, our advanced security application would open certain doors. The light sensor would let our application know if lights are on in the rooms that the recognized visitor has access to, and turn them on prior to this person entering them.

Our design allows IoT data to be sent from the Raspberry Pi to the cloud. We could use this data in a remote dashboard app. A remote security agent could provide an override if, for example, our application could not make a match of a legitimate visitor.

Let's now expand on the key benefits, as these relate to our advanced security application.