- By Kudzai Manditereza
- July 11, 2024
- HiveMQ
- Feature
Summary
Exploring design principles, data and functional modeling, and security considerations for architecting a unified namespace.
Manufacturers are increasingly leveraging digital technology to drive business outcomes through data-driven intelligence. However, a significant challenge remains: The consolidation of diverse data from various operational technology (OT) and information technology (IT) systems into a central repository for meaningful analysis. This traditional approach often fails due to inconsistencies and complexities in data from varied sources.
A more modern approach is unified namespace (UNS), which externalizes contextualized data into a real-time semantic hierarchy, establishing a hub-and-spoke model that serves as a single source of truth (Figure 1). This article explores practical guidance for architecting a UNS, covering design principles, data and functional modeling, and security considerations.
Principles of a unified namespace architecture
These principles are foundational to the successful implementation of a UNS architecture.
Edge-driven and domain ownership. The UNS architecture relies on edge-driven data propagation, where components within specific functional domains push data from its source into a common data infrastructure. This edge-driven approach, using a report-by-exception model, ensures that data is shared only when changes are detected, creating a real-time representation of the enterprise’s state and events.
Domain ownership allows each functional area to package data according to its unique semantics, while adhering to federated data governance guidelines. This autonomy enables tailored data models for specific advanced analytics use cases, empowering regular users to analyze data without relying on central data scientists.
Open architecture and standard data infrastructure. An open architecture using standardized methods for information exchange is crucial for a UNS. The MQTT protocol is the preferred communication standard, offering efficient, reliable and decoupled data sharing through a publish/subscribe model.
Open architecture fosters innovation by allowing teams to access necessary tools without navigating technical compatibility issues, avoiding the accumulation of technical debt. It also ensures future adaptability to technological advances without overhauling existing infrastructure.
Federated data governance. Federated data governance manages data quality and ensures interoperability within a distributed, domain-oriented environment. It promotes the development of universal standards and policies while honoring the governance of individual domains. This approach balances localized control with overarching harmony across all domains, empowering teams close to the data to manage and share it effectively.
Core components of a UNS
These core components are essential to building a robust unified namespace for efficient data management.
MQTT broker for the unified namespace. The MQTT broker is the central hub for data communication in a UNS architecture. Multiple MQTT brokers, tailored to specific needs, may be used, such as HiveMQ for high availability and HiveMQ Edge for machine connectivity. Ensuring full adherence to OASIS standard MQTT specifications is critical for maintaining an open architecture.
IIoT platform for the unified namespace. An Industrial Internet of Things (IIoT) platform bridges the gap between modern and legacy equipment, facilitating data collection from diverse sources and publishing it to the MQTT network. It organizes and refines data, enhancing its readability and reliability, and ensuring it is prepared for integration and analysis across different domains.
The combination of an MQTT broker and an IIoT platform forms the core of the UNS, ensuring that new data is shared across the network and that localized or immediate data requirements can be met.
Data persistence for the unified namespace. Data persistence is crucial for an effective UNS architecture, requiring both a historian (time-series database) and a structured database (SQL). The historian archives data over time, allowing for retrospective analysis, while the SQL database holds structured data for operational management and analysis.
The historian mirrors the live data model of the UNS, providing a historical record accessible through the IIoT platform. Transactional operations on stored data may be facilitated through the IIoT platform, bridging the UNS and the data store.
Unified namespace reference architecture with MQTT
The UNS reference architecture is based on MQTT brokers and IIoT platforms (Figure 2), with additional tools like HiveMQ Edge for data conversion and integration. Data from devices with MQTT capabilities and connectivity solutions like KepserverEX is fed into a “raw MQTT namespace,” then processed and redistributed into the UNS.
MQTT bridges enable multiple brokers and levels of a manufacturing enterprise to share information, creating a distributed network that ensures data consistency and availability across the network.
Designing your UNS semantic information hierarchy
A clear semantic hierarchy is essential for effective data access and integration. This hierarchical organization, structured into layers, provides a single source of truth, reflecting the business’s current state and events.
UNS semantic data hierarchy design using MQTT. In an MQTT-based system, the broker organizes data using a topic hierarchy, acting as a structured framework for data access. This hierarchy (Figure 3) enables precise control over data sources within the UNS, allowing participants to efficiently access required data.
The structured nature of MQTT topics enables a comprehensive data access hierarchy within an organization. The hierarchy typically follows the ISA-95 common data model, which reflects the organizational hierarchy within the MQTT topic structure.
Best practices for MQTT topic namespace structuring for UNS
Adopting the ISA-95 common data model helps in organizing the MQTT topic namespace, reflecting the enterprise’s structure. This structure ensures that information about data access is available independent of the actual data content, maintaining consistency and organization within the UNS.
By adhering to these principles and best practices, manufacturers can successfully implement a UNS, facilitating seamless data integration, enhancing data-driven decision-making and driving digital transformation.
About The Author
Kudzai Manditereza is an experienced technology communicator and electrical engineer based in Germany. As a developer advocate for HiveMQ, his goals include creating compelling content to help developers and architects adopt MQTT and HiveMQ for their IIoT projects. Manditereza also runs a YouTube channel and podcast covering IIoT and smart manufacturing technologies. He has been recognized as one of the Top 100 global influential personas talking about Industry 4.0 online.
Did you enjoy this great article?
Check out our free e-newsletters to read more great articles..
Subscribe