The Industrial Internet of Things (IIoT) is often associated with big data and clouds, collecting large amounts of data from widely distributed sensors and turning "information into insight." In some industrial processes, insight time is critical, and the delay in sending data to the cloud and receiving responses can be too long. In other cases, data security may be affected, or a fast, reliable connection to the Internet may not be available. To overcome these challenges, edge computing can complement the big data processing power of cloud computing. It performs compute-intensive tasks that require immediate response and stores and filters data into the cloud when appropriate. Edge computing can include simple elements such as data filtering, event processing, and even machine learning, and can be hosted on any connected device, from a small embedded computer or PLC to an industrial PC or local micro data center. Isolate and take up very little space from other processes running on the same platform is a key requirement.
Maximize the value of your data
Much of the value of IIoT is that it can bring together information from multiple sources to help companies see a bigger picture: how to improve processes, optimize maintenance activities, reduce waste and energy consumption, and eliminate avoidable costs. The cloud-centric IIoT view shows the various data streams that are aggregated and analyzed in a remote data center using heavyweight software applications.
The premise of this model is that reliable Internet connections are always available, with enough bandwidth to handle the data pushed to the cloud, and this delay - from data generation to receipt of cloud feedback results is acceptable. However, any of these important ingredients may be lost. Remote sites may rely on cellular networks for Internet connectivity, but coverage may be incomplete or unreliable. A large number of sensors can generate large amounts of data that are costly to communicate with the cloud, especially if they contain high-definition images or video. Complex decisions may be required in real time for security reasons or to maintain operational efficiency. On the other hand, for some companies, data security can be a problem because these companies may not want to pass sensitive information to the cloud over the Internet.
In any of these cases, it may be impractical to send raw data captured from a process or device back to the cloud. Nonetheless, some intelligence and decision-making capabilities are required at the individual machine or process control logic level to enable companies to determine the best course of action. This is the role of edge calculations.