principal33 | Hybrid and Multi-Cloud in German Utilities: How to Decide Which Workloads Go to Azure, AWS or Private Cloud Skip to main content

The right question is not “which cloud”, it is “which workload”

In the cloud debate, German utilities are often pushed into a binary decision between vendors: Azure or AWS, public or private, hyperscaler or sovereign. That is the wrong question. The right question is which workload should go where, because a utility is not a single system but a constellation of applications with very different risk profiles, regulatory criticality and consumption patterns.

The context reinforces this logic. According to the Bundesnetzagentur Monitoring Report 2025, German utilities already operate in a market with 54% renewable generation, 25 million meter locations using modern metering equipment, 644,000 mandatory smart metering systems installed and new obligations on controllable loads under section 14a EnWG (264,874 market locations notified by the end of 2024). This operational diversity makes it impossible to compress the entire technology stack into a single cloud decision. Bundesnetzagentur.

What CTOs Really Worry About – And How to Make Complex Tech Transitions Simple

Four workload profiles, four different decisions

In a typical utility, workloads fall into four profiles. The first are market and customer-facing platforms (portals, CRM, digital service, Salesforce). These typically benefit from hyperscalers such as Azure or AWS for elasticity, integration with Power Platform or managed services, and the maturity of their European regions.

The second profile consists of critical regulated systems (SAP IS-U, FI-CA, billing, market communication). Here the priority is not elasticity but stability, traceability and sovereignty. Many utilities choose private cloud or hyperscaler in European regions with documented BSI C5 controls.

The third profile is grid operational systems (SCADA, asset management, substation control). For many operators these workloads remain on-premise or on private cloud due to KRITIS and latency requirements, although the analytics layer on top of their data can live in cloud.

The fourth profile is analytics and AI workloads (forecasting, customer analytics, predictive maintenance). These benefit from public cloud thanks to on-demand compute and ML service ecosystems, always with clear controls on where data resides and is processed.

BSI C5 as a transversal reference

A point that saves many discussions is anchoring decisions in the BSI Cloud Computing Compliance Criteria Catalogue (C5). The BSI publishes C5 catalogues as control criteria for cloud services, especially relevant in sectors with KRITIS requirements. Using C5 as an objective reference to evaluate providers and configurations replaces preference-based discussions with an auditable matrix. BSI.

Why hybrid usually wins in utilities

Operational reality means that most German utilities end up in a hybrid and often multi-cloud architecture. Not as a trend, but for three concrete reasons. First, legacy does not disappear: SAP IS-U and other systems remain the functional core and cannot be replaced overnight. Second, regulatory and sovereignty risks push some workloads to highly controlled environments. Third, innovation — Data & AI, customer experience, smart metering analytics — requires cloud capabilities that only hyperscalers offer at scale.

The frequent mistake is not going hybrid, but reaching hybrid by accumulation, with no design. That generates silos, fragile integrations and rising operating cost. A real hybrid strategy requires explicit decisions on which workload lives where, how environments connect, how identities and data are managed and how team fragmentation is avoided.

Where Principal33 fits

Principal33 includes Cloud Migration & Support and Application Maintenance & Support among its core services and states specific experience in demanding sectors such as energy. In its content on SAP IS-U, it also highlights hybrid transition models for companies not ready to move everything to S/4HANA, and modular rollout plans. This fits the workload-based decision logic: coexist with legacy while progressively modernising what brings most value or most risk. Principal33.

principal33 | Hybrid and Multi-Cloud in German Utilities: How to Decide Which Workloads Go to Azure, AWS or Private Cloud