Intelligent edge: The future of cloud computing?

The intelligent edge refers to the collection and analysis of data where the data is generated. It could be the point where a user such as a mobile worker is doing their job, or an IoT device attached to industrial equipment is generating data. The data is collected, processed and analysed at this remote point, meaning the intelligence sits on the ‘edge’ of the organisation’s IT architecture. This ‘decentralisation’ of storing data, processing and intelligence capabilities takes much of the workload away from the central processing hub, datacentre or cloud.

Edge computing is a powerful enabler when it’s injected with intelligence to enhance its potential for disruption and enrich the value of the data captured at its proverbial source, says Scott Cowling, director, network solutions at BT. “The sheer volume of data created through the IoT, where information is streamed from sensors to optimise operations, makes this intelligence a business imperative.”

For Hanno Brink, machine learning engineer at Synthesis, the technical challenge that the intelligent edge is trying to solve is the efficient use of resources, increasing robustness, and reducing cost of implementation and management of this technology.

“There are many benefits that could be unlocked if this technology is applied correctly. One is the reduction in latency between when data is collected and acted upon, which has applications such as predictive maintenance, fleet maintenance, and many others where real-time visibility and responsiveness would aid in reducing operating costs or improving service delivery.

“Another would be that intelligent applications delivered to customers where limited bandwidth or connectivity has previously prevented such, or where data from customers hasn’t been easily collected and operationalised. This technology could also be used to collect data that wasn’t previously accessible, due to the cost of implementation, and to provide new customer experiences or improve current service delivery by using intelligence where the customer is interacting with it.”

Forrester’s 2021 ‘Build integrated technology platforms to accelerate growth and agility’ report tells IT leaders: “In this stage of technology maturity, innovate with technology to serve customers across the entire customer lifecycle, mastering systems of engagement and insight.”

This is precisely what the intelligent edge allows us to do, says Varsha Ramesar, managing executive, data and analytics at iOCO. “From a technology perspective, the intelligent edge has several benefits, including reduced dependency on network performance, and allows the business to increase its bottom line by reducing overhead expenses. What organisations must bear in mind, though, is that the real value of the intelligent edge lies in how it can expand and amplify a company’s ability to sense and respond with greater speed and agility – whether that’s in the context of predictive maintenance, or customer service.”

Navinder Singh, GM at In2IT Technologies, says edge computing emerged in the past few years driven by a few technology companies that developed the technology. When the intelligent cloud arrived, it was a major innovation as it helped to overcome some of the fundamental issues of cloud computing such as latency, bandwidth requirements and cost containment. In the past year, cloud providers have been accelerating intelligent cloud offerings, where data that sits on the edge with intelligence built into it can still interact with, and relay, to and from the cloud.

In addition, he says investment into the edge has changed its potential and scope. “If we look at the intelligence incorporated into edge devices, we’re seeing advancements that align with cloud strategies and one of the most challenging elements – integration with the cloud. The intelligence is creating seamless integration of applications into the cloud through AI. For instance, companies are often compelled to integrate their CRM and ERP solutions with cloud service providers, yet there’s a lot of effort required from resources for these applications to achieve seamless integration. AI gathers information from the edge applications and as the integration requirements are predefined, this simplifies the integration process significantly and, as a result, reduces costs. It also eliminates bottlenecks where processing and analysis was conducted centrally with data alignment across multiple applications. Processing becomes faster and more efficient.”

Moving data and compute closer to the edge also means that this infrastructure is moving further away from your control, and it becomes vastly more exposed.

Hanno Brink, Synthesis

However, like any other technology that is, relatively speaking, in its infancy, the intelligent edge is not without its challenges. Brink says these include creating more efficient hardware, creating more compute-efficient AI, looking at more efficient ways to manage limited bandwidth and storage resources, securing the edge, distributing machine learning to the edge to preserve privacy, and overcoming myriad challenges presented by implementing systems that interact with the real world.

Various proof-of-concept and proof-of-value projects have proven the ROI and business benefits of intelligent edge deployments that bring AI and ML to edge environments, adds Ramesar. However, the challenge now is determining how to scale these deployments to hundreds or thousands of sites so that organisations can take full advantage of the business-critical data they are generating at the edge. Before businesses can realise the benefits of the edge and embark on their industrial digital transformation, they must consider their data. The volume and velocity of data is growing astronomically, and availability is key. To support these applications and use cases, sensor and related contextual data must be ingested, processed, and analysed in the right place, at the right time and provided to the right people.

For Cowling, one of the main concerns at the edge, and one of the biggest barriers to the deployment, is cybersecurity. Increasingly the edge becomes a point of convergence between two worlds: operational technology (OT), including industrial systems that run equipment in factories, refineries and mines, and IT. Industry 4.0 solutions like predictive maintenance need data from both worlds, such as SCADA from OT and ERP from IT.The OT world is highly vulnerable to cyber-attacks, the vast majority of which come from IT and has traditionally relied on security through obscurity and air gaps. “Once you join the dots, critical processes become vulnerable because they run on old proprietary software, with poor password protection, limited patching, and no authentication. Identifying and mitigating vulnerabilities therefore becomes a major focus area.”

When it comes to securing the intelligent edge, Brink says the most crucial component of any edge application is making sure that your application is secured from the device, all the way up to the cloud, and back again. “Moving data and compute closer to the edge also means that this infrastructure is moving further away from your control, and it becomes more exposed. Any such application must be designed with security top of mind.

“Over and above data security, physical security and reliability also become factors that must be considered and appropriately balanced with functionality. In addition, secure physical devices should be used, and these devices should be robust against the extreme environments that they may operate in.”

Organisations should make sure that systems can scale effectively while also ensuring a consistent security implementation, and need to define an organisational root of trust. This is a way for edge devices to authenticate themselves to the company, and prevents privileged systems from being impersonated and their access abused.

Security and privacy risks can be reduced by limiting data flows between the point of collection and the core infrastructure, particularly when those flows happen over the public internet, adds Cowling. Using the intelligent edge helps businesses to adhere to in-country data protection laws. It keeps sensitive data within the device, anonymising and analysing at source rather than sending identifiable information to the cloud.

Secure access service edge, or SASE, is also key to bringing together connectivity and network security into a single policy-driven service that provides consistent, centrally managed access and security from edge-to-edge. SASE also supports a zero trust approach to the cloud and underlying infrastructure, meaning sessions are protected regardless of where the edge device is connecting from.

As we move closer to cloud and hybrid cloud environments, the benefits that the intelligent edge deliver make it a compelling technology procurement decision.

Navinder Singh, In2IT

Securing the edge is not just daunting, it can seem downright impossible, considering the unprecedented number of devices on the network, generating data every second of the day, all of which needs to be ingested, transformed, and analysed by compute platforms in the wild, and all needing to be locked down.

“Surprisingly, many IT professionals think of security as securing perimeters and implementing robust access control. However, security in today's era of edge and cloud is so much more complicated. With the explosion of IoT, and Industrial IoT in particular, the attack surface has increased, as has the number of attack vectors. Since edge computing is a distributed model, its security concerns are very different from a centralised model. To save costs and speed up deployment, many edge devices don’t encrypt data natively, and IT managers need a security framework in place before the large-scale rollout of edge projects,” says Ramesar.

Ultimately, as we move closer to cloud and hybrid cloud environments, the benefits that the intelligent edge deliver make it a compelling technology procurement decision, says Singh. It helps in streamlining the business processes, data alignment across multiple applications, easy integrations and more. It’s also all about the business optimising its operations and driving efficiency, and in an era of ‘always on’ and ‘instant gratification’, the intelligent edge is the way to go.

* This feature was first published in the May edition of ITWeb's Brainstorm magazine.