What is PRP?
PRP stands for Parallel Redundancy Protocol and is a network protocol that provides seamless network redundancy by transmitting packets simultaneously over two independent networks. PRP is standardised in IEC 62439-3.
PRP is used in industrial environments where zero packet loss, maximum availability and determinism are required — such as in energy, chemicals, rail and critical process automation.
🧠 How does PRP work?
- Each device with PRP (a DANP – Dual Attached Node PRP) is connected to two physical networks: LAN A and LAN B
- Each network packet is sent twice – once via each network
- The receiving node accepts the first packet that arrives and ignores the duplicate
- Both networks are fully active and independent
- If one network fails, communication remains fully intact and without delay
PRP provides zero recovery time: no failover is needed — communication remains active at all times.
🏭 Use of PRP in industrial networks
- Critical infrastructures such as power grids, nuclear power plants and rail traffic
- Networks with strict requirements for availability and real-time performance
- Dual redundant networks combined with SCADA, PLCs and Drives
- Used together with protocols such as IEC 61850, OPC UA and ProfiNET
- Supported by industrial devices from vendors such as Hirschmann, Siemens, ABB and Ruggedcom
PRP is particularly suitable for zones 0 and 1 in the Purdue Model.
🔍 PRP vs. MRP vs. RSTP
| Aspect | PRP (Parallel Redundancy Protocol) | MRP (Media Redundancy Protocol) | RSTP (Rapid Spanning Tree Protocol) |
|---|---|---|---|
| Recovery time | 0 ms (no interruption) | <200 ms | 1–10 sec |
| Topology | Two fully separate networks | Ring | Flexible (mesh, ring, tree) |
| Complexity | High | Moderate | Low |
| Use in OT | For critical, loss-free communication | Common standard for redundancy | Basic failover in standard networks |
🔐 Security considerations
- PRP increases availability, but not necessarily security
- Both networks must be secured separately with Firewall, ACL, VLAN and SIEM
- Monitoring must take place on both channels to detect anomalous behaviour
- Use RBAC and physical security on all network components
- Avoid cross-connecting switches to prevent errors or loops
Segmentation and consistent management of both PRP networks are essential for safety and reliability.
📌 In summary
PRP provides unmatched network redundancy without delay or data loss, essential for business-critical industrial applications. Although more complex than other redundancy methods, it is the only solution that guarantees zero-time failover.