Use Cases
EU AI Act · NIS2 · GDPR · Clean Energy Package
Energy: Cross-operator data sharing for critical grid infrastructure
Grid operators are now classified as critical infrastructure under the NIS2 (Network and Information Security Directive). Energy data, from smart meters to cross-border grid flows, must be shared across operators while each retains control of their own systems. Apeiro's federated architecture makes this possible without a single central platform.
The challenge today
Smart meters deployed across Europe generate data that is personal data under GDPR; it reveals occupancy patterns, appliance use, and daily routines. The infrastructure managing this data typically sits on commercial cloud platforms whose legal structure may not satisfy the strictest data protection requirements.
At the same time, transmission system operators managing cross-border energy flows must share real-time data for grid balancing, but are now classified as essential entities under the NIS2 directive. A security incident affecting one operator’s systems can cascade. NIS2 fines for essential entities are substantial, and the supervisory obligations are significant.
The EU’s Clean Energy Package requires cooperation across operators, countries, and regulatory frameworks. But the data-sharing infrastructure connecting these operators is fragmented. Each cross-border integration is typically a bespoke bilateral arrangement. The result is that pan-European optimisation of renewable energy dispatch and demand forecasting remains far harder than the technology would otherwise allow.
What becomes possible with Apeiro
IronCore and CobaltCore provide the infrastructure layer within each operator’s own facilities, with bare-metal provisioning and IaaS (Infrastructure as a Service) managed through Kubernetes-native APIs, extended audit capabilities, and customer telemetry services built into CobaltCore. Each operator’s systems remain under their own jurisdiction, operated under EU law, with the security baseline NIS2 requires for essential entities. Greenhouse’s operational platform runs on top, providing the compliance dashboard, policy violation monitoring through its DOOP (Decentralised Observer of Policy Violations) plugin, and Prometheus-based observability. Incident detection and reporting obligations are met as a continuous operational output, not a separate compliance exercise.
Cross-operator data sharing is handled without a central platform. ORD (Open Resource Discovery) lets each operator describe their available data feeds (grid sensor streams, generation forecasts, meter aggregates) as machine-readable metadata. The UMS (Unified Metadata Service) aggregates these descriptions across operators, so participants can discover what is available and under what conditions without bespoke bilateral integration work. Platform Mesh’s federated service marketplace provides the runtime layer: each operator publishes their data services; other operators discover and subscribe to them through standardised interfaces, each retaining full control of their own infrastructure throughout.
Because each Gardener-managed cluster can be deployed to target EU-certified infrastructure, compute-intensive workloads such as optimisation models and demand forecasting algorithms can be scheduled to data centres running on renewable energy, making sustainability targets an operational parameter rather than an aspiration.
The AI systems running these forecasting and grid-optimisation models fall under the EU AI Act as high-risk systems operating in critical infrastructure. OCM (Open Component Model) records each model version as a signed artifact, capturing the training framework configuration, model code version, and infrastructure snapshot, giving grid operators a tamper-evident record of the software stack the Act requires to be documented. The Knowledge Graph, built over ORD-described data products from participating operators, provides AI agents with a semantic map of available grid data across the network, enabling cross-operator optimisation queries without manual data-sharing agreements for each model update.
A scenario
Three national transmission system operators want to share real-time demand forecasting data to optimise cross-border load balancing during high renewable generation periods. Each operator’s infrastructure runs on IronCore with Greenhouse providing the NIS2-compliant operational layer. Each operator publishes their data feeds via ORD; the Unified Metadata Service makes these discoverable across the group. Platform Mesh handles subscription and lifecycle: each operator’s data stays in their own infrastructure and is shared under machine-readable governance policies; no third-party platform holds a copy. When a fourth operator wants to join, they connect to the same standardised layer rather than building new bilateral integrations with each of the three existing participants.
Apeiro is an open reference architecture — a collection of components and blueprints. Realising this scenario requires industry-specific knowledge, engineering investment, and integration expertise. Explore the technical foundations
