Cloud to Edge

IoT & Edge Platform

Run every site from one console, even when the link drops.

IoT & Edge Platform extends Neor out of the data centre and onto the sites where operations actually happen: substations, factory floors, hospital wings, branches and remote field locations. It is a single Kubernetes-native control plane that reaches every edge cluster, connected device and inference workload you run, managed from a console you host inside your own jurisdiction. Each site runs autonomously, so applications keep executing and policies keep applying when the link back to the core is down.

Edge estates usually grow faster than the operating model around them: another vendor, another operating system, another management console, and a fresh round of field engineering for every site added. That overhead grows with the estate, and it lands on the same budget that has to prove, at every audit, that operational data never left the country. Consolidating onto one control plane makes adding a site a configuration change rather than a project, and leaves data residency enforced by the architecture instead of asserted by a process.

Architecture

How it is put together

Central Control Plane

A single management cluster in your own data centre holds fleet state, policy, the application catalogue, identity and the audit record for every site you operate.

Cloud-Edge Messaging Fabric

An encrypted, ordered, store-and-forward channel carries instructions down and telemetry up, buffering at both ends through outages and reconciling automatically on reconnect.

Edge Workloads & AI Inference

Containerised applications and validated models are distributed to sites and run locally on whatever hardware is present there, from constrained gateways to GPU-equipped site servers.

Edge Cluster Runtime

A lightweight Kubernetes-native runtime at each site schedules workloads, enforces policy and persists data locally, with no dependency on the core to keep running.

Device & Protocol Layer

Sensors, controllers and OT equipment are onboarded over standard industrial protocols and presented to applications in a normalised, addressable form.

Capabilities

Zero-Touch Site Onboarding

A new edge node registers, collects its policy set and reaches a managed, auditable state from one console, regardless of vendor or operating system.

Offline Autonomy

Nodes continue scheduling workloads, enforcing policy and writing data locally when connectivity is unavailable, so a network outage does not halt work at the site.

Store-and-Forward Sync

Encrypted, ordered message delivery between core and site, and directly between sites, with state synchronised automatically once the link returns.

Fleet Application Delivery

Deploy containerised applications to a single node or an entire fleet using Kubernetes-native scheduling, with versions and rollouts controlled centrally through GitOps.

Device & Protocol Connectivity

Connect OT and IT equipment over standard industrial protocols without maintaining a separate integration gateway stack at every site.

Edge AI Inference

Run models locally for computer vision, anomaly detection and predictive maintenance, so sensitive operational data is never routed to an external inference endpoint.

Central Policy & Audit

Define data-handling, network and access policy once at the centre; the platform enforces it at every node and records what ran where.

Mixed-Fleet Hardware Support

x86 and ARM nodes, from small gateways to GPU-equipped site servers, are managed as one estate under the same lifecycle and policy.

Where it fits

A grid operator runs protection and telemetry workloads in substations where the backhaul drops for hours and the site has to keep operating on its own.
A hospital analyses imaging on-site so patient data is processed inside the clinical network and never crosses to an external endpoint.
A manufacturer converges OT and IT on the factory floor and runs predictive-maintenance models next to the machines they monitor.
A bank or retailer keeps branches and stores trading through a network outage, processing transactions locally and settling centrally when the link returns.
A national operator manages remote field sites such as pipelines, wells and pumping stations from one console, sending engineers out for hardware faults only.

What you end up with

The customer ends up owning a single, auditable edge estate: every site onboarded, governed and kept running from one control plane inside their own borders, with an audit record showing that operational data stayed there.