Ring + MRP for production
Ring topology with MRP is the standard for industrial automation. Fast failover, simple cabling.
9.4 Network Topologies in OT
The previous chapter covered the standards that govern industrial networks. Those standards define protocols and security requirements, but they do not dictate how switches connect physically. Topology choices are driven by physical constraints, redundancy requirements, and protocol support.
Why Topology Matters in OT
Section titled “Why Topology Matters in OT”In IT networks, a brief outage is an inconvenience. In OT networks, a 200 ms outage causes a PLC watchdog timeout and stops production. The topology determines how quickly the network recovers from a failure, and whether it recovers at all.
Star Topology
Section titled “Star Topology”A central switch connects to all other switches. Simple and easy to manage. The core switch is a single point of failure. Mitigate with redundant core switches and LACP or RSTP.
Use star topology for office areas, small machine cells, and non-critical applications.
Ring Topology
Section titled “Ring Topology”Switches connect in a closed loop. One link is blocked under normal operation. When a link fails, the blocked link opens and traffic flows the other way around the ring. MRP provides recovery in less than 200 ms.
Ring topology is the most common in industrial automation: production lines, conveyor systems, and any application requiring fast redundancy.
Daisy-Chain Topology
Section titled “Daisy-Chain Topology”Switches connect in a line with no loop. No redundancy. A single link failure splits the network. Simple and low-cost.
Use daisy-chain for simple machine cells, temporary installations, and non-critical monitoring.
The choice between these topologies depends on the scale of the plant. Large plants combine them into a hierarchical architecture.
Hierarchical Cell Architecture
Section titled “Hierarchical Cell Architecture”Modern industrial networks follow a hierarchical cell architecture aligned with the Purdue Model. Each cell (Level 1) is a ring of switches serving one production area. Cells connect to the plant network (Level 2 to 3) through a distribution switch or firewall.
Topology Selection Guide
Section titled “Topology Selection Guide”| Requirement | Recommended Topology |
|---|---|
| Fast failover (< 200 ms) | Ring with MRP |
| Maximum redundancy | Redundant star with RSTP/LACP |
| Simple, no redundancy needed | Daisy-chain or star |
| Large multi-cell plant | Hierarchical (rings at cell level, redundant star at plant level) |
| Substation (IEC 61850) | HSR or PRP |
Key Takeaways
Section titled “Key Takeaways”Hierarchical for large plants
Use rings at the cell level and a redundant star at the plant level. Align with the Purdue Model.
Avoid daisy-chain for critical systems
Daisy-chain has no redundancy. A single link failure splits the network.
What Comes Next
Section titled “What Comes Next”The topology defines the physical layout. The protocols that run on that topology define how devices exchange data in real time. The next chapter covers the Industrial Ethernet protocols: PROFINET, EtherNet/IP, HSR/PRP, and TSN.
References
Section titled “References”- IEC 62443-3-2:2020 — Security risk assessment for system design
- Hirschmann. (2023). Network Design Guide for Industrial Ethernet. Belden/Hirschmann.
- PROFIBUS & PROFINET International. (2020). PROFINET Network Topology Guide.