What is Spanning Tree Protocol? How It Works

The Spanning Tree Protocol, which is sometimes just called “Spacing Tree,” is like Waze or MapQuest for modern Ethernet networks. It sends traffic down the most efficient route based on what’s happening right now.

Spanning Tree is based on an algorithm that American computer scientist Radia Perlman made for Digital Equipment Corporation (DEC) in 1985. Its main goal is to stop redundant links and communication paths from looping in complex network configurations. As a secondary function, Spanning Tree can route packets around trouble spots to make sure that communication can flow through networks that may be having problems.

Difference between Spanning Tree Topology & Ring Topology

In the 1980s, when companies first started to connect their computers together, the ring network was one of the most common ways to do it. For example, IBM’s Token Ring technology was first used in 1985.

In a ring network topology, each node is connected to two others: one that is in front of it on the ring and one that is behind it. Signals can only go around the ring in one direction, and at each node along the way, packets are passed on to the next node.

When there are only a few computers in a network, simple ring networks work well. However, when there are hundreds or thousands of computers in a network, the rings don’t work as well. A computer might have to send packets through hundreds of nodes just to share information with another system in the next room. Bandwidth and throughput are also a problem when traffic can only go in one direction and there is no backup plan in case a node along the way breaks or gets too busy.

In the 1990s, as Ethernet got faster (100Mbit/sec was introduced in 1995) and the cost of an Ethernet network (bridges, switches, and cables) got much cheaper than Token Ring, Spanning Tree won the LAN topology wars and Token Ring quickly died out.

How Spanning Tree Protocol Works?

Spanning Tree is a protocol for sending data packets where they need to go. It’s like being both a traffic cop and a civil engineer for the roads that data travels on. It sits at Layer 2, which is the data link layer, so it only cares about getting packets to where they need to go. It doesn’t care what kind of packets are being sent or what data is in them.

Spanning Tree is so common now that its use is written into the IEEE 802.1D standard for networking. Based on what the standard says, there can only be one active path between any two endpoints or stations for them to work.

Spanning Tree is made to make it impossible for data moving between different parts of a network to get stuck in a loop. In general, loops mess up the forwarding algorithm that is built into network devices. This means that the device no longer knows where to send packets. This can cause frames to be sent twice or duplicate packets to be sent to more than one place. Messages can be sent over and over. The person who sends a message can get it back. It can even crash a network if there are too many loops. This is because loops use up bandwidth without gaining much, and they stop other traffic from getting through.

Loops can’t form because the Spanning Tree Protocol closes off all but one path for each data packet. Switches in a network use Spanning Tree to define root paths and bridges where data can travel and to close off duplicate paths, making them inactive and unusable while a primary path is open.

Because of this, network communications run smoothly, no matter how big or complicated a network gets. In a way, Spanning Tree uses software to make single paths through a network for data to travel. This is similar to how network engineers used hardware to make single paths through the old loop networks.

Additional Benefits of STP

The main reason why Spanning Tree is used is to stop routing loops from happening in a network. But there are also other good things about it.

Spanning Tree is able to tell if a node on one of the primary paths has been turned off because it is always looking for and defining which network paths can be used to send data packets. This can happen for a number of reasons, from a broken piece of hardware to a change in how the network is set up. It could even be a short-term problem caused by bandwidth or something else.

When Spanning Tree finds out that a primary path is no longer being used, it can quickly open a previously closed path. It can then send data around the trouble spot, making the detour the new primary path or sending packets back to the original bridge if it becomes available again.

The original Spanning Tree made these new connections fairly quickly, but in 2001, the IEEE came out with the Rapid Spanning Tree Protocol (RSTP). RSTP, which is also called the 802.1w version of the protocol, was made to make recovery from network changes, temporary outages, or the complete failure of components much faster.

And while RSTP added new ways for paths to converge and new roles for bridge ports to speed up the process, it was also made to work with the original Spanning Tree. So devices that have both versions of the protocol can work together on the same network.

Flaws of the Spanning Tree Protocol

Even though Spanning Tree has been around for a long time and is used by almost everyone, there are some who say it’s time has come. The biggest problem with Spanning Tree is that it closes off possible paths for data to move through a network, which can lead to loops. About 40% of the possible network paths in any network that uses Spanning Tree are closed off to data.

In very complicated networking environments, like those found in data centres, the ability to quickly scale up to meet demand is essential. If Spanning Tree didn’t have so many restrictions, data centres could open up a lot more bandwidth without having to buy more networking hardware. It’s kind of funny that this is happening because Spanning Tree was made for situations like this. And now, the protocol’s protection against looping is, in a way, keeping these environments from being as good as they could be.

Multiple-Instance Spanning Tree (MSTP), a better version of the protocol, was made so that virtual LANs could be used and more network paths could be open at the same time without loops forming. But even when MSTP is used, there are still a lot of possible data paths that can’t be used on a network.

Over the years, there have been a lot of different, non-standardized attempts to improve the bandwidth limits of Spanning Tree. Even though the people who made some of them say they worked, most of them aren’t fully compatible with the core protocol. This means that organisations have to either use the non-standard changes on all of their devices or find a way to make them work with switches that use standard Spanning Tree. Most of the time, it’s not worth the cost to keep and support multiple versions of Spanning Tree.

Future of STP

Aside from the fact that Spanning Tree closes network paths, which limits bandwidth, there hasn’t been much thought or work put into replacing the protocol. Even though IEEE sometimes releases updates to try to make the protocol more efficient, they are always compatible with the version that is already in use.

Spanning Tree is kind of like the saying, “If it ain’t broke, don’t fix it.” Most networks have Spanning Tree running in the background to keep traffic moving, stop crash-causing loops from forming, and route traffic around trouble spots so that end users don’t even know if their network is temporarily down as part of its normal operation. On the back end, network administrators can add new devices without worrying too much about whether or not they can talk to the rest of the network or the outside world.

Because of all this, it’s likely that people will keep using Spanning Tree for a long time. There may be some small changes now and then, but the core Spanning Tree Protocol and all of the important things it does are likely to stay.