A UK SME called Iceotope is building a prototype security system for the small, distributed data centres that will soon process most of the world’s data. These “edge” facilities—think server rooms in lampposts or apartment blocks—are proliferating because of 5G and video streaming. But they face a fundamental security gap: the computers (IT) and the cooling/power systems (OT) do not talk to each other, creating vulnerabilities and wasting energy. The project, MoatE, will build a Secure Edge Facility Orchestrator (SEFO) that uses the Arm Morello platform’s hardware-based security compartments to give operators a unified, hardened view of both IT and OT networks. If successful, the demonstrator, due by 2025, could cut CO₂ emissions from edge computing and make smart-city infrastructure—traffic lights, power grids, public Wi-Fi—more resilient to cyber-attacks. It also tackles a practical problem: without this IT-OT bridge, the efficiency gains of edge computing could be lost, returning to the dismal power-usage ratios of pre-2000s data centres. The project is applied, not fundamental science, and its impact is squarely on infrastructure that most people never see but increasingly rely on.
View original technical description
MoatE - A cyber-security business demonstrator project for Edge Computing: * Led by UK SME Iceotope; * The University of Manchester providing technical expertise; * Avnet providing the voice of the customer, and end-user evaluation for the project's demonstrator. * Avnet will also offer global channels for dissemination/demonstration and future sales when subsequently productised. Gartner predicts that by 2025, 75% of data will be processed at-the-Edge, driven by: 5G, latency, bandwidth limitations, video streaming, and user-generated content. Edge computing facilities will be small and distributed (e.g. many 10kW-200kW vs. few 60MW datacenters) leading to: * Potential return to power inefficiency; prior to consolidation in 2000's of datacenters, common Power Usage Effectiveness (PUE) were dismal \>2 * Increased attack-surface, both virtual and physical * Increased complexity for maintenance and asset-management * Space issues, e.g., derived from new Edge facilities in densely populated areas Iceotope has a key enabling technology enabling density and improved efficiency. However, there is a cyber-security barrier for implementing effective Edge computing: a lack of cooperation between Information Technology (IT) (e.g. servers), and Operational Technology (OT) (e.g. the cooling, facility and power). This lack of cooperation reduces key cyber-security risks for these "symbiotic" IT and OT systems but exacerbates the inefficiencies of Edge facilities. Iceotope has been active in working with key industry standards bodies (Redfish and ASHRAE) to begin to address this, but the cyber-security barriers remain. An opportunity exists for taking advantage of the new capabilities of the Morello platform to address this IT and OT cooperation barrier, whilst also demonstrating substantial value in key areas of: * Cyber-Security (virtual & physical); * Resilience; * Efficiency; * Effectiveness; * Maintenance (proactive, predictive and reactive). MoatE will develop a Secure Edge Facility Orchestrator (SEFO) that provides a unified and secure view of information across networks (IT and OT) at Edge facilities, harnessing the new security compartmentalisation features of the Morello platform. SEFO will be further hardened against cyber-attacks and implement information analysis and mechanisms to deliver improvements for fault detection/prediction, efficiency-optimisation, asset management (including foreign asset detection); and detection of security attacks. MoatE will deliver a Morello-based demonstrator for Edge-facilities by 2025, showing added value in: cyber-security; resilience; operational management; and efficiency. Evidencing reductions in CO2 emissions, and added social value in enabling and securing the smart cities of the future.
Plain English summaries and category classifications on this site are generated by AI and may not perfectly reflect the original research.
Is something wrong? Let us know