Network Redundancy Function of Managed Industrial Switch

Network redundancy is a method of preventing network outages. More specifically, maintaining a redundant network requires installing copies of various network components to ensure that in the event of a failure of the original network component, a backup can seamlessly replace it. If you create a copy of each component, you can keep the network up and running 24/7/365 because if the network goes down all you have to do is switch to the copy. It's called a "redundant" network because the best-case scenario is that your network never experiences downtime. So the ideal situation is to have redundant (more AKA) backup copies of all the components that make up the network.

Using a managed industrial switch over an unmanaged industrial switch,one of the benefits of this allows you to use Ethernet with an extra connection, so if one path between two points on the network fails, the other path can be used to deliver messages. If one link or industrial switch fails, another link or industrial switch can take over to prevent unnecessary downtime.

So why not physically connect the unmanaged industrial switches in the network in different loop configurations so that there are always at least two paths between each industrial switch?

Because such a connection would create a broadcast loop that would quickly paralyze the network. In an unmanaged Ethernet network, there can only be one path between any two ports on the network. If there are multiple paths from one unmanaged industrial switch to another, broadcast messages sent by the network (and in some cases other messages) will be forwarded and complete the loop on the return to the second path. Since unmanaged industrial switches forward all broadcasts and do not keep track of the messages they send, the returned messages will be sent again and again in a loop. It is obviously not a good thing for a message to be transmitted back and forth at high speed in a loop, so unmanaged industrial switches do not allow looping.

The limitations of having only one link are even clearer. If the only path fails for any reason, such as a broken cable or a power failure in one of the industrial switches, there is no transmission path and no network traffic to go through. So we need a way to add alternate paths without creating loops. Use redundancy protocols such as RSTP (Loop Prevention Protocol) so that managed industrial switches can communicate with each other to detect and prevent loops.

RSTP: Rapid Spanning Tree Protocol Rapid Spanning Tree Protocol is currently the preferred method for purposefully creating rings that allow multiple redundant paths on a network. This feature allows managed industrial switches to intelligently decide a path when the network comes up and assign a backup path if some part of the original path fails. How it decides the original path and changes to an alternate path is much faster than when using the original spanning tree protocol. Enabling legacy STP is only really useful if your legacy network requires this protocol. The RSTP feature is usually enabled by default.

In actual networking, you can see the advantages of managed industrial switches for network transmission. When an aggregation industrial switch is connected to 3 designated industrial switches, each designated managed industrial switch is connected to each other. If this is an unmanaged industrial switch, creating such a redundant network will directly cause the unmanaged network to collapse. But using the redundancy protocol of RSTP, the managed industrial switch will temporarily block the backup path so that the loop will not interfere with the network.

When the active path between the aggregation industrial switch and the managed industrial switch is disconnected, the interconnected backup path is activated so that no industrial switch is lost from the network. In an unmanaged network, a path failure removes the industrial switch from the network, making it impossible for all end devices on the industrial switch to communicate.

Therefore, in order to minimize network latency and improve the deterministic level of the network, we must first choose managed industrial switches.