Network administrators often use network topology diagrams to understand where to best place nodes and links, but they aren’t necessarily starting from scratch. There are several types of network topologies, including: point-to-point topology, bus topology, ring topology, star topology, tree topology, mesh topology, and hybrid topology.

A point-to-point network, or a point topology, is the easiest network to understand, and the most basic type of network topology. It’s simply two nodes that are connected by a single link. Data travels back and forth between these two endpoints. While this is the easiest network type to set up, its simplicity is its own drawback. A point-to-point topology isn’t applicable for most modern use cases.

In a bus network, each node is connected to a single cable, like bus stops branching off from a bus route. The data transmission all flows through that one central connection. Because everything is connected in a straight line off one central cable, it’s a cost-effective topology, and simple to set up and add new nodes.

However, a shared central link does have drawbacks. In a system that has one central link with many dependencies, a failure of that central link causes all dependencies to fail. Bus networks are also not as secure as some other types due to this shared central link. Additionally, the more nodes that share a central cable, the slower a network is.

In a ring network, the nodes and links are arranged in a ring. Each node has exactly two neighbors. In such a network, repeaters are used to ensure that data can reach the nodes that are farthest away from each other in the ring. Data usually flows unidirectionally in a ring network.

These kinds of networks are cheap to install and expand, and data flows quickly within the network. But a failure of a single node can bring down the whole network. Dual-ring networks are used to protect against this type of failure.

A dual-ring network features two concentric rings instead of one, and the rings send data in opposite directions. The second ring is used when there is failure in the first ring, and this type of network is often used to support critical infrastructure.

In a star network, all the nodes are connected to a central hub. The nodes are positioned around that central hub in a shape that roughly resembles a star.

This topology type makes it easier to troubleshoot issues with a particular node. If a single node fails, the rest of the network is unaffected. That said, if the central hub goes down, then the entire network goes down with it. In a star network, the performance of the entire network depends on the central hub and the connections to it.

It’s useful to think of a tree topology as a combination of a bus network and a star network. In a tree topology, you still have one central hub that is connecting everything, but instead of single nodes that branch out from that central root node, it’s other star networks. This topology allows for more devices to connect to a central data center, which speeds the flow of data. Like in a star network, identifying issues with individual nodes is relatively easy.

Tree topologies have the same drawbacks of both bus and star networks, namely the vulnerability to a single point of failure. If that central connection goes down, everything goes down.

In a mesh network, each device is connected to at least one other node in the network. In a full mesh network, each node is connected to every other node. In a partial mesh network, only some of the nodes connect directly with each other, while others need to pass through additional nodes to reach the targeted node.

Since nodes can communicate directly with each other, rather than through a central hub, communication on a mesh network is often very fast. A great example of a mesh network is the internet itself, where each computer is a node in a network provided by different internet service providers that also connect to each other. Mesh providers like NYC Mesh use a mesh network to provide wireless internet across one of America’s most populous cities.

Because mesh networks have multiple routes over which information can travel, they are more resilient than many other topologies and can continue to function if a node or connection fails. Mesh networks also offer greater security—if a node is attacked or compromised it can be replaced.

However, mesh networks are often expensive to set up, requiring many cables to establish the connections. Meshing multiple paths can also make installation complicated and lead to maintenance costs that are higher than in other types of topologies.

A hybrid network topology is any type of network that uses a combination of topologies. A tree network that combines a star network and a bus network is a type of hybrid topology.

Hybrid networks offer flexibility and help organizations design a topology that specifically meets its needs. However, creating a customized network architecture can be challenging and can require more cabling and network devices, raising maintenance costs.