The redundant title of this article is your first clue to the topic of this article: PROFINET media redundancy.
PROFINET media redundancy is optional with all conformance classes, but in some applications, it is critical for correct operation (for example, hot backup in a safety application). PROFINET offers network media redundancy through a ring topology. PROFINET has two main media redundancy classes defined. MRP – Media Redundancy Protocol which offers a few milliseconds of recovery time in a network up to 50 switches and MRPD – Media Redundancy for Planned Duplication which is a bumpless redundancy. (Zero millisecond recovery time)! MRP is typically used with Real Time (RT) and MRPD with Isochronous Real Time (IRT) but you can interchange depending on application requirements.
MRP and MRPD use the design of a ring manager which monitors the ring topology at all times for proper operation. If the ring becomes ‘broken’ then the manager has to notify the other ‘clients’ on the ring that there is a failure and then switch to a line topology. In the case of MRP, this process may take a few milliseconds, but MRPD is faster and bumpless as it sends real-time frames already in both directions on the ring to the recipient of the data.
You can learn more about these in our MinutePROFINET video series. Here’s a short animation of how they work: