In its IIoT context, the edge is traditionally understood to be a kind of half-way point between the plant and the cloud. Physically hard to define, the edge acts as a repository for data collected from the plant’s smart sensors and actuators, prior to sending it on to cloud-based analytic applications for processing and interpretation.
The cloud, while also physically hard to define, is conceptually better understood by non-IT personnel. It’s a place away from the plant, a server farm for instance, where masses of computing power and data storage capacity are available on a for hire basis. The potential drawbacks with cloud-centric strategies in a processing environment usually have little to do with the cost though, but involve things like latency, bandwidth, OT protocol support and information security.
The question then becomes, which data is better processed in the cloud, and which at the edge? In some cases the answer is clear cut, in others not so much. For instance, it’s unlikely that anyone would risk putting critical safety related applications in the cloud just yet, for reasons to do with latency and security. On the other hand, matters relating to energy efficiency and environmental monitoring may be ideal for cloud-based implementation due to their less stringent real-time and security requirements. I’m not suggesting environmental disasters of themselves cannot be catastrophic, just that the early warning signs are usually detectable long before things actually become critical, which means latency of even minutes is probably acceptable here. An example of a monitoring application that could easily be hosted in the cloud is the level in a dam, say, which only needs to be monitored once per day.
The edge then is a conduit to the cloud, but it can also act as a data processing destination in itself, in line with the mantra: “edge where you can; cloud where you can’t”. This evolving “duality” in the nature of the edge has given rise to the terms “thick-edge” and “thin-edge”. Thick-edge refers to architectures where data processing happens at the edge (vs in the cloud), while thin-edge refers to an architecture optimised to transfer large amounts of data to and from the cloud.
In these refined edge-to-cloud setups, industrial Ethernet switches are expected to remain firmly rooted in their connectivity role at the thin edge of the emerging multi-tiered stack, with continued reliance on thick edge devices for edge computing. See the ARC Advisory Group article on page 38 for more on how decentralised, autonomous plant operation can be facilitated through distributed communication and advanced network infrastructure.
All the best for 2020
On behalf of the team at SA Instrumentation and Control, welcome back to all our readers and advertisers – 2020 is going to be another challenging year and we wish you all the very best in your endeavours.
Steven Meyer
Editor: SA Instrumentation & Control
| Tel: | +27 11 543 5800 |
| Email: | malckey@technews.co.za |
| www: | www.technews.co.za |
| Articles: | More information and articles about Technews Publishing (SA Instrumentation & Control) |
© Technews Publishing (Pty) Ltd | All Rights Reserved
printer friendly version