The Arrowhead Tools project, funded by the German Federal Ministry of Education and Research under grant 16ESE0359, delivered an Extended Historian Service (EHS) that serves as a core data‑collection component for the Arrowhead Framework. The service was designed to meet the needs of industrial data analytics, energy efficiency, task management, and smart maintenance. It provides secure, encrypted communication between data sources, the historian, and analysis tools, and is intentionally simple for analysts to configure and use.
The EHS is built on a gRPC‑based architecture that supports both standard remote procedure calls and streaming calls. The service uses SSL/TLS encryption and mutual authentication via certificates that are signed by a system administrator’s self‑signed root. The administrator creates a public/private key pair, signs certificates for the EHS server and its adapters, and the server and clients verify each other’s credentials at runtime. This security model is fully documented in the open‑source repository (MIT licence, https://github.com/ifak‑prototypes).
From a technical standpoint, the EHS offers a Python API that allows analysts to retrieve current sensor values or historical time series with minimal configuration. Analysts only need to specify the EHS IP address and provide the paths to the certificate files. The API exposes an EHSClient class that reads a configuration file and exposes methods such as `get_current_value()` and `get_time_series()`. This design reduces the learning curve and enables rapid integration into existing data‑analysis pipelines.
Performance testing was carried out on both Linux (virtualised with Oracle VirtualBox) and Windows 10 machines. The tests measured the time per data transfer for varying payload sizes and numbers of consecutive service calls. Results showed that the transfer time decreases as more calls are bundled together, confirming the benefit of batching. On Windows, the payload size had a stronger influence on transfer time than on Linux, yet overall call times were lower on Windows. Importantly, the streaming mode of gRPC—where a single client request can elicit multiple sequential responses from the server—consistently outperformed the simple RPC mode, indicating that streaming is the preferred approach for high‑throughput installations. A streaming variant of the adapter was implemented and verified, though further work is needed to improve software stability for production use.
The project was organised into several work packages. IFAK contributed to WP1 (requirements and state of the art), WP3 (core system design and implementation), and WP5 (configuration component). In WP1, IFAK helped define the use case “Production Support, Energy Efficiency, Task Management, Data Analytics and Smart Maintenance” and established the interoperability requirements for data acquisition. WP3 saw IFAK design the EHS architecture and implement the core service, while WP5 focused on the configuration interface that allows analysts to set up connections to the historian.
The collaboration involved multiple partners within the Arrowhead Tools consortium, though the report lists only IFAK explicitly. The project ran until March 2023, with iterative development cycles that first delivered core functionality and later focused on performance optimisation. The final deliverable is an open‑source, MIT‑licensed EHS that can be integrated into the Arrowhead Framework, providing secure, efficient, and user‑friendly data collection for industrial analytics.