If you want to know about what to do with sensitive data in operational technology (OT) systems – how to prioritise and segment it, where to place it between the edge and the cloud, and how to retrieve it at both ends (and anywhere between) – then you could do worse than ask Siemens. Which is what RCR Wireless did, in pursuit of definitive answers about the ‘critical edge’, in particular as it plays with new private (and possibly public) 5G systems.
What follows is a transcript of an email interview with Marc Fischer, the company’s global marketing manager for the industrial edge, and Yiyao Fu, presented as an ‘industrial 5G expert’ at the Germany-based industrial giant. Their answers will be threaded into a new editorial report on the critical 5G edge, out later this month, which is also attended by a webinar next week (September 27) on the same subject; you can sign up to the webinar here.
…
How should one rationalise usage of edge and cloud (and in-between) compute resources for processing critical Industry 4.0 workloads?
Fischer: “When considering IoT/OT workloads, there are key factors to consider, as follow:
– Latency | Edge computing reduces delays, while cloud computing may introduce latency through data transmission.
– Bandwidth | Edge computing conserves bandwidth by processing data locally, ideal for limited network capacity.
– Privacy / security | Edge computing keeps sensitive data closer, reducing exposure during transmission.
– Resources | Edge computing eases strain on limited cloud resources, useful for restricted connectivity or computational capabilities.
– Operation | Edge computing allows applications to function without the internet, which is crucial when constant cloud connection is uncertain.
“While the decision may vary depending on industry requirements, a hybrid approach that combines both edge and cloud computing can ultimately achieve the desired outcomes for critical IoT/OT workloads. But very sensitive or critical industries such as pharmaceuticals or public infrastructure providers typically seek to keep all or most of the data on premise to maintain the high IT security requirements – also important, given their legal constraints.
“By contrast, in non-sensitive consumer goods industries – such as warehouses or electronics [production], for example – it is often the case that certain applications may run more or less equally on the premises or in the cloud; it just depends on the company’s individual strategies.”
Is there a rule-of-thumb for when IoT / OT data should be retained at the on-site edge, and when it can be diverted by the cloud?
Fischer: “There is no rule of thumb. Companies need to carefully evaluate the key factors, as outlined above. It depends on the real-time or near-time requirements of the applications – whether they run directly at the machine or in the private / public cloud. If you look at a computer vision application, for example, which tracks the quality of processes or produced goods, then the AI runs directly near the machine as it is closely connected to the PLC and camera information.
“But, typically the training of these AI algorithms is done in the cloud, where nearly unlimited storage and processing power is available. The below matrix shows which use cases / applications (in which industries) will be handled at the edge and in the cloud.”
Are there Industry 4.0 scenarios where the data will stay on the edge, all the time, separately of the public internet? Can you give examples?
Fischer: “Yes, as follow:
– Machine-to-machine communication in single production networks in sensitive companies;
– AI-based applications that require real/near-time information (machine vision, anomaly detection etc);
– Automation control in connected applications such as virtual PLCs, closed loop AI applications, virtualized sensors etc;
– Connectivity / interaction with local MES or SCADA systems;
– Additional information on top of a human-machine interface (HMI), such as development of forces and anomalies, and so on – which is to be available to the station or machine operator only.”
Will all-edge private 5G networks, where cellular is required, always be used in these cases?
Fu: “A private 5G network can be operated with or without an internet connection. Enterprises need to make that decision themselves. The advantage of private 5G networks is that companies know exactly where the 5G system is hosted and can assure no external parties can access private data or intellectual property within the network. Of course, when it comes to edge or cloud, private 5G networks are only the transport mechanism for the data within the space – and public 5G or wireless LAN (Wi-Fi) or Ethernet are all alternatives.”
Can you say (again, rule-of-thumb / on paper) how latency reduces with 5G from the cloud to the edge, in terms of networking and computing, when employed separately and together?
Fu: “It is a combination of both – the ‘best of both worlds’. For low latency, 5G is much better than 4G-LTE, say. But 5G and edge/cloud dov not necessarily correlate – it depends on how the customer uses the technologies and configures the network. 5G is just one of the technologies that can be used to connect the edge to the cloud.”
All of this, presumably, makes clear how complex Industry 4.0 deployments can be, and how important ecosystem consultancy and collaboration is. Is this right? Can you comment?
“Ecosystems are very important. Industry 4.0 deployments can be complex. 5G can be used to facilitate deployments [in a different way] than previous mobile standards. Our own private 5G infrastructure is designed as an open ecosystem, which allows connection of third-party devices. We will offer support, testing and certification for those devices to ensure a smooth operation on the shop-floor. Furthermore, our 5G routers can be run on third party 5G networks, both private and public. The industry needs interoperability to make sure that devices on the shop-floor interact smoothly with each other. So interoperability is essential for us.”
For more on this topic, tune in to the upcoming webinar on Critical 5G Edge Workloads on September 27 – with ABI Research, Kyndryl, Southern California Edison, and Volt Active Data.
All entries in the Postcards from the Edge series are available below.
Postcards from the edge | Compute is critical, 5G is useful (sometimes) – says NTT
Postcards from the edge | Cloud is (quite) secure, edge is not (always) – says Factry
Postcards from the edge | Rules-of-thumb for critical Industry 4.0 workloads – by Kyndryl
Postcards from the edge | No single recipe for Industry 4.0 success – says PwC
Postcards from the edge | Ultra (‘six nines’) reliability – and why it’s madness (Reader Forum)
Postcards from the edge | Private 5G is reshaping the Industry 4.0 edge, says Nokia
Postcards from the edge | Something strange at the 5G edge, says Southern California Edison
