One of the biggest challenges in establishing the Industrial Internet of Things (IIoT) will be seamlessly Internet-enabling the 'Things' that live at the edge of the network. Industry-wide, this area contains trillions of Things that contain one or many data points that may need to be analysed and combined into information.
For IIoT to be successful, several challenges need to be overcome, including the ability to:
In order to alleviate some of these challenges, IIoT strategies will focus on pushing data into a centralised cloud platform. This platform and its corresponding services will be administered by IT experts familiar with the managed world of IP and made available to anyone with the proper credentials and an Internet connection. Leveraging the power of cloud computing and its multitudinous resources will make the required storage and processing power available to handle the zettabytes of data that will be collected, analysed and archived. Furthermore, the overall uptime of these platforms continues to trend upwards as they become more resilient to the increasing demand and expectations of our connected world.
The actual source of data pushed into the cloud resides within the Industrial Things that live at the edge of the network. The edge bridges the gap between IT and Operational Technology (OT), where the rich resources available in the cloud are not directly available. OT encompasses industrial networks that have their own nuances and introduce additional challenges.
Very often, industrial networking technologies do not leverage Ethernet as their physical communications layer. Depending on the environment and the Things that comprise a system, anything from RS232/485 to modems or proprietary wiring may be encountered. Likewise, the data protocols that are exposed over these communication mediums are not likely to be IP derivatives. Consequently, a hodge-podge of industrial networks has been created with no attention paid to the future possibility of being connected to the Internet.
Unlike IP addresses in the IT world, many industrial Things have their own addressable schemes for uniquely identifying themselves on the network. These schemes vary by vendor and type, and may not have built-in discovery mechanisms. Innate knowledge by an integration expert is required to interconnect the Things in a way that makes them function as a whole.
Industrial Things have historically followed a request/response model. If a particular Thing is interested in a piece of data contained in another Thing, it will make an appropriate connection, request the piece of data and wait for a response containing the result. Although this pull model is fine for Things living within the same digital boundary of OT, security and scalability requirements will dictate that this model is unacceptable for the outside IT world trying to look inbound. Instead, IIoT prefers a push model, where industrial data flows outbound to a cloud platform.
In order to seamlessly integrate industrial data into IIoT, a new communications platform is required. This platform requires extensive knowledge of the intricate realm of OT and the state-of-the-art and rapidly changing domain of IT.
Within OT, the platform must understand the various network topologies and data protocols that will be encountered. It must be able to automatically discover and identify industrial Things and the data they contain, as well as be able to handle the storage of high frequency updates.
Within IT, the platform must be able to transform the data it collects and push it into the cloud via IIoT standards. Emerging standards include Asynchronous Messaging Queuing Protocol (AMQP), Message Queuing Telemetry Transport (MQTT), Constrained Application Protocol (CoAP), and Data Description Services (DDS). These standards allow for the retransmission of data in the event it does not reach its destination.
With the lack of computer networking infrastructure in OT, this platform must be embeddable and run within a standalone appliance or an edge-based switch or router where IT and OT converge. Its flexibility will enable industrial data to be sampled cyclically or based on some event or condition and be published to the cloud independently of data collection. Data filtering should be available through basic analytics. Lastly, user setup should be minimal by automating as much configuration as possible.
As industry continues to define IIoT, the concepts and realisation of the optimal Embedded IIoT solution will continue to evolve.