The rapid convergence of Information Technology (IT) and Operational Technology (OT) is forcing industrial enterprises to modernize their legacy infrastructures. However, integrating decades-old machinery with modern cloud platforms creates significant architectural and security challenges, requiring a standardized approach to industrial IoT design.
Key layers of industrial IoT architecture
Integrating IT and OT (Operational Technology) environments requires a standardized approach to data architecture. According to the AWS Well-Architected IoT Lens guidelines, system reliability is established during the design phase of the entire data pipeline, which consists of the following layers:
- Physical layer: sensors that generate signals.
- Controller layer (PLC): signal acquisition for real-time process control.
- Edge layer: industrial gateways for initial filtering and protocol conversion.
- Data transmission layer: message brokers for telemetry delivery.
- Platform layer: cloud or on-premises time-series databases (Time Series DB) for storage and analysis.
Legacy equipment integration and the role of OPC UA
The primary challenge for enterprises is legacy equipment operating on proprietary protocols without built-in encryption. To unify data, industrial Edge gateways and the OPC UA (OPC Unified Architecture) standard are utilized. It ensures platform-independent transmission of not only numerical values but also metadata (data type, timestamp, quality status), enabling the integration of legacy machines into modern IT systems.
Distributed processing: Edge computing vs. cloud
Sending all raw data directly to the cloud overloads communication channels. At the Edge device level, initial filtering and telemetry analysis are performed close to the source. Cloud resources are reserved for long-term historical data storage, predictive maintenance, and building digital twins.
Operational technology (OT) security
According to the NIST SP 800-82 guidelines, system availability takes precedence over data confidentiality in industrial networks. Security relies on strict network segmentation to isolate production equipment from the corporate IT environment. The ISA/IEC 62443 series serves as the international standard for managing these risks.
Infrastructure technological maturity levels
- Level 0 (Chaotic): data is collected manually or isolated within local SCADA systems.
- Level 1 (Basic monitoring): basic telemetry is available without a unified transmission standard.
- Level 2 (Standardized): use of OPC UA, Edge devices, and IT/OT network segmentation.
- Level 3 (Integrated): real-time data visualization on dashboards, integrated with MES/ERP.
- Level 4 (Predictive): implementation of predictive maintenance scenarios and digital twins.
Implications for business
Failing to establish a robust IoT architecture leads to fragmented data silos, increased vulnerability to cyber threats, and high bandwidth costs from transmitting raw data. Conversely, successful integration enables predictive maintenance, reduces operational downtime, and allows enterprises to transition from manual monitoring to fully automated, data-driven decision-making.
Recommendations
- Assess maturity: Evaluate your current infrastructure against the five technological maturity levels to identify gaps.
- Implement Edge computing: Deploy industrial gateways to filter and process telemetry locally before cloud transmission.
- Standardize protocols: Adopt OPC UA to unify data formats across legacy and modern equipment.
- Segment networks: Isolate production environments from corporate networks following ISA/IEC 62443 standards to ensure OT security.
Prepared by a Software Ukraine member. Original publication.