The Future of Big Data

Data Handling for Software-Defined Military Systems

Software-defined systems (SDS) represent a significant advancement from traditional military platforms to those controlled primarily by software. This transformation enables advanced digital capabilities, increases operational security, and networks systems intelligently with each other and the broader infrastructure. SDS remain current throughout their lifecycle through Over-The-Air (OTA) updates. However, the exponentially growing volume of data poses a significant challenge for data management.

Solutions for Current and Future Data Management Challenges

Data volumes increase both within complex onboard systems and through growing connectivity with external sensors, command structures, and other platforms. Particularly in military development and operational testing, these data volumes demand advanced analytics and fault diagnosis capabilities. “Modern defense platforms are highly complex due to their software-driven architecture and diverse assistance and autonomy functions. Therefore, it is necessary to efficiently store, manage, and analyze data—often in the petabyte range. To meet these challenges, we developed IAV Merida,” explains Dr. Remo Lachmann, data expert at IAV. This SaaS solution simplifies handling large data volumes during the development and operational lifecycle of military systems.

Data as a Foundation for Successful Operational Testing

IAV Merida receives data from various sources: data loggers integrated into vehicles, drones, or unmanned systems, as well as mobile telemetry units. These devices specifically record relevant operational and mission data and transmit it to the platform. This data is crucial for fault identification and evaluation of system behavior under real operational conditions. “Based on this data, we identify where new networking functions fail or where the platform shows unexpected behavior,” says Lachmann. Before deployment in real operations, systems undergo extensive field and lab testing to verify quality and reliability. Regardless of climatic or terrain conditions, each test generates vast measurement data that are efficiently stored and analyzed.

Storage and Processing of Massive Data Volumes

Extensive testing causes IAV Merida to store hundreds of terabytes of operational and test data monthly. Overall, the platform processes multiple petabytes per month, as each measurement is subjected to various analyses. Data is currently predominantly held in secure, on-premises data centers, but the platform can flexibly run in cloud environments (e.g., Azure, Tencent) – tailored to the respective security requirements.

“Our customers always retain full control over their sensitive data, whether on-premises or in the cloud,” emphasizes Dr. Lachmann.

Worldwide, Secure Data Management and Compliance

To operate in different regions, IAV Merida complies with all local data protection and security regulations. For example, in China, local cloud providers are used to meet strict regulations such as the export ban on geodata. This ensures the full functionality even in security-critical environments.

Automated Analytics and Interactive Dashboards

IAV Merida performs over 300,000 automated analyses monthly to extract precise insights from complex measurement data. These are visualized in interactive dashboards featuring charts, tables, statistics, and geographic maps. The analyses enable real-time monitoring and rapid response to anomalies during ongoing tests or operations—both at the individual vehicle and fleet level.

Integration of Artificial Intelligence and AI-Based Anomaly Detection

Beyond classical and rule-based analytics, IAV Merida employs AI algorithms to detect unknown failure patterns and behaviors that are difficult to find with conventional methods. The AI-based anomaly detection generates a predictive score that identifies potential failures and defects early.

“With AI anomaly detection in IAV Merida, we can identify and fix potential problems weeks or months before their actual occurrence,” says Dr. Lachmann.

Dr. Remo Lachmann

Head of Department for Data Analytics