Manufacturing
Germany
SMA manages a globally distributed fleet of inverters, battery systems, and prosumers in their energy management platform that continuously generate high-volume telemetry and event data, including status, performance, faults, grid conditions, and forecasts. This data must be ingested, processed, and analyzed in near real-time to support use cases such as predictive maintenance, anomaly detection, fleet optimization, and advanced reporting for internal teams and external customers.
The customer's primary objective is to improve its energy solutions business by leveraging operational and field data from inverters, storage systems, and prosumers in their energy management platform.
Cloudera was chosen over other platforms due to its robust hybrid capabilities. Since 2020, SMA has been working with Cloudera on-prem but recognizes the potential in future to train advanced analytics and machine-learning models in the cloud.
This approach facilitates real-time anomaly and performance issue detection, more reliable operation of photovoltaic and storage systems, and more efficient service processes, ultimately enhancing availability, reducing costs, and increasing the value of SMA’s digital energy services.
Use Cases, Solution and Impact
This data is used to enable predictive maintenance, detect anomalies, and optimize plant and fleet performance, while also providing a comprehensive view of system health and energy flows to operations and service teams.
Cloudera Data Platform: This platform provides a central, secure data foundation where SMA can store and manage large volumes of historical and real-time data with low total ownership costs.
It serves as the central analytics and governance layer, supporting use cases like model training, self-service analytics, and dashboards for operations, service, and management.
Cloudera Data Streaming: It offers a scalable streaming backbone for ingesting and processing continuous data flows from SMA’s devices in their energy management platform.
It supports real-time anomaly detection, alerting, and event-driven workflows, allowing field issues to be detected and addressed with minimal delay.
Story developed in May 2026
