Redundant rings explained: RSTP, MRP, ERPS, PRP and HSR
KINNEX Team5 min read
When a cable is cut or a switch fails, a redundant network sends traffic the other way. How quickly it does so depends on the protocol.
RSTP
Rapid spanning tree is a general-purpose standard that prevents loops and recovers within a second or so. It is widely supported and fine for non-critical traffic, but slow for control networks.
MRP
Media redundancy protocol, common in PROFINET networks. A manager switch watches the ring and recovers in tens of milliseconds, making it well suited to automation.
ERPS
Ethernet ring protection switching, defined by ITU-T G.8032. Fast ring recovery, used in industrial and carrier networks.
PRP
Parallel redundancy protocol sends every frame over two independent networks at once. If one path fails, the other already carries the data, so recovery time is zero. It needs two parallel networks and suitable end devices.
HSR
High-availability seamless redundancy applies the same idea to a ring: frames travel both ways around it. Used in substations and power automation.
Choosing
- Standard office-style loops: RSTP.
- PROFINET or general automation: MRP.
- Multi-vendor industrial rings: ERPS where supported.
- Zero-recovery needs, such as protection in substations: PRP or HSR.
Test, do not assume
Whichever protocol you pick, validate it on the bench or on site: break links, power cycle switches and measure what happens to the real traffic. Record the results in the test report.