The computing edge is becoming more powerful. From the smartphone in our pockets to the desktop (laptop) devices we use every day and outward to the airport check-in ‘kiosk’ computers we use… and across to the digitally-linked traffic lights that we pass through with carefree oblivion, our machines are all getting smarter. As we know, much of this intelligence is now being driven by Artificial Intelligence (predictive, reactive and of course generative) and its ability to process information through pattern recognition processes that enable us to to live our lives better.
While users’ wider notions of AI in these scenarios is that the intelligence is happening on the devices themselves and through the pipe of juice that cloud datacenters provide us with on the backend, an important estate of data exists on more remote Internet of Things (IoT) devices, which are now being empowered with enough edge computing engine power to perform a significant degree of their own algorithmic intelligence.
Requirements for Industry 4.0
Often thought of as the birth of so-called Industry 4.0 (steam, electrified production and computing being the first three), analyst house McKinsey defines this period in our evolution as, “The next phase in the digitization of the manufacturing sector, driven by disruptive trends including the rise of data and connectivity, analytics, human-machine interaction and improvements in robotics.” All well and good then, but we can’t just plug these machines in and expect them to work, they need a new approach to high bandwidth data services that offer seamless connectivity and – even when they are out on the computing edge – they need to perform with secure low latency connections.
With reasonably balanced pedigrees in this space for some years now, NTT Data and Schneider Electric have come forward with a new data fusion co-innovation initiative as a means of providing Industry 4.0 devices, networks and infrastructure with the elements they need to create our future smart factories, to build our smart cities and do the next-big-thing in smartness that will inevitably come next.
Gen-AI at the edge
The companies describe their new joint work effort as an initiative designed to integrate edge computing assets and devices with private 5G technologies, IoT and modular datacenters with the aim of creating connectivity and with a view to supporting the computational demands of generative AI applications deployed at the edge. For completeness here, let’s define modular datacenters (MDCs) as a way of building typically smaller datacenters (often housed in something similar to a shipping container or a pre-fabricated building) with hardware and software components that are architected and engineered to be, well, modular i.e. more easily capable of being added to or reduced depending upon the needs of an installation at any given facility.
Known for its work in digital technologies as well as energy management and automation, Schneider Electric will now integrate with NTT DATA’s Edge-as-a-Service (as it sounds, edge computing services for IoT devices, but provided as a managed, controllable & supported service), which includes fully managed Edge-to-Cloud, private 5G and IoT capabilities. The fusion point is where NTT DATA’s Edge-as-a-Service meets Schneider Electric’s EcoStruxure, a modular datacenter that combines Operational Technology (OT – hardware & software to monitor devices, processes and infrastructure for data processing and networking reasons) with other technologies.
AI inference difference
What all this brings to Industry 4.0 deployments is power, but it’s a specific kind of power designed to handle compute-intensive tasks such as machine vision, predictive maintenance and other AI inferencing applications at the edge.
“We’ve listened to our customers and know that processing vast amounts of data generated by edge devices is where the future of digital transformation lies. That’s why we’re excited to announce that we have the solution to meet these obstacles and are ready to lead the way towards a more connected and efficient digital world. We are thrilled at the prospect of what our continued collaboration with Schneider Electric means for the industrial and other sectors,” said Shahid Ahmed, EVP of new ventures and Innovation at NTT Ltd.
This union of technologies is designed to enable the deployment of edge datacenters tailored in remote and brownfield locations, where high compute demands infrastructure such as power, cooling, racks and specialized IoT and AI management systems. As previously explained here, brownfield sites are already muddy i.e. the deployment is done on a partly ploughed field where existing infrastructure exists in some form to build on but where there are still legacy data migration tasks, a variety of integration tasks and new seeds (software code) to sow.
As companies look to use edge compute to support automation and enable data-driven decision-making, NTT Data points to its Edge Advantage report which suggests that nearly 70% of enterprises are accelerating edge adoption to solve critical business challenges. The first manifestation of a private 5G enabled deployment of an EcoStruxure datacenter is at Marienpark, a Berlin-located innovation zone that spans over 74 acres.
“After leveraging NTT Data’s expertise in private 5G connectivity and then maximizing synergies with our EcoStruxure architecture in our facilities, it’s time to expand our collaboration and bring a complete solution to industrial customers,” said Rob McKernan, president, cloud & service provider segment at Schneider Electric. “Together, we aim to assist global clients in adopting connected devices, specialized industrial solutions and the right edge computing infrastructure with modular datacenters to gain valuable data insights, particularly in the context of IoT and emerging AI requirements.”
Building stronger edge (computing)
Camille Mendler, chief analyst, enterprise services, Omdia reminds that AI-enriched data already accounts for a third of enterprise network traffic. “But looking forward, it will dominate digital interactions by 2030,” said Mendler. “To profit from AI insights, enterprises must invest in digital resources at the edge and the technology infrastructure that powers it, now. Industry 4.0 relies on actionable, data-driven intelligence delivered in real-time whether that be in a factory, an industrial park, an airport or an office campus,” she said.
What’s involved here involves a lot of deep-tech infrastructure and network backbone technologies, but what’s happening here is simple, we need to see edge computing (i.e. those computations, analytics and data management processes that happen outwise on the Internet of Things) given more power.
No, robot arms aren’t going to suddenly start thinking for themselves, but actually yes, they are to a degree… as they start to use generative AI functions to understand what’s happening next on the production line, whether they need maintenance and what time tea break is. Okay, robots don’t go on tea (or coffee break), but you get the point – the explosion of AI applications means we need more power on the computing periphery and the perimeter of these devices itself is brought closer to the core enterprise application and stack.
