Network Topologies

Network topologies are the physical and logical arrangements of devices in a network that define how these devices are connected and communicate with each other. The history of network topology can be traced back to the 18th century, when the Königsberg bridge problem inspired Swiss mathematician Leonhard Euler to develop network topologies. This concept has evolved considerably over time, particularly with the advent of computer networks in the late 1960s and early 1970s, which were characterized by the development of the Arpanet.

There are various basic types of network topologies, which differ in the arrangement of devices and have various advantages and disadvantages. The most common topologies are:

Star topology

Concept: In a star topology, all nodes (computers, printers, etc.) are connected to a central hub or switch. Each node has its own connection to this central device.

Advantages: Easy to set up and manage. Adding or removing devices is simple and does not affect the rest of the network. Fault detection is easier as each device is connected to a central hub.

Disadvantages: The central hub is a single point of failure. If it fails, the entire network can be affected.

Example: A home Wi-Fi network in which several devices such as laptops, smartphones and smart TVs are wirelessly connected to a central router.

Bus topology

Concept: In the bus topology, all nodes are connected with a single backbone cable. The data sent from a node is transmitted in both directions until it reaches its destination node.

Advantages: Requires less cable length than other topologies and is easy to implement for small networks.

Disadvantages: Limited in size and scope. The entire network can become inoperable if the backbone cable fails or if too many devices are connected, resulting in signal degradation.

Example: Old Ethernet networks often used a bus topology where the computers were connected with a thin coaxial cable with BNC connectors.

Ring topology

Concept: The nodes in a ring topology are connected to each other in a circle. Each node is connected to two other nodes and forms a ring. The data moves in one direction and is forwarded from one node to another until it reaches its destination.

Advantages: As the data is transmitted in one direction, there are no data collisions. It is also relatively easy to recognize and isolate errors.

Disadvantages: The failure of a single node or connection can disrupt the entire network, unless it is a dual ring or bypass technology.

Example: Some old classroom computer labs may have used a ring topology for networked computers, where each computer was connected to its neighboring counterpart.

Tree topology

Concept: Starting from a root node, this hierarchical structure branches out through various levels of sub-nodes. The topology combines elements of star and bus topology. Star-shaped groups of devices are connected by a backbone.

Advantages: The simple expansion by adding branches and levels to the network enables high scalability. As all information runs via the central node, errors can be easily identified.

Disadvantages: Failure of the backbone cable can bring the entire network to a standstill. The topology is also susceptible to attacks, as compromised devices can potentially access all data. 

Examples: Examples of the use of tree topology can be found in computer networks, especially in home or small office environments where multiple devices such as printers and computers need to be connected, often creating bus networks around a main server.

Mesh topology

Concept: Each node in a mesh topology is connected to every other node. There are potentially two types of mesh topologies: Full Mesh, where every node is connected to every other node, and Partial Mesh, where nodes are selectively connected to each other.

Advantages: Provides high reliability and redundancy. If a node fails, alternative paths can be used for data transmission.

Disadvantages: Complex configuration and management due to the large number of connections. It is also expensive due to the large number of cables and connections required.

Example: The Internet itself can be seen as an example of a mesh topology, with multiple paths between any two nodes in the network.

Hybrid topology

Concept: a hybrid topology is a combination of two or more different topology types. For example, a large office building could use a star topology for each floor but connect the floors together using a ring topology.

Advantages: Extremely flexible, as components of different topologies can be integrated to best suit the requirements at hand.

Disadvantages: The design can be complex and costly, depending on which topologies are combined.

Example: A company network in which the individual departments are organized in a star topology, while the departmental networks for communication between the departments are connected in a ring or bus.

Choosing the right topology for an industrial fieldbus or Ethernet network requires careful consideration of several factors to ensure efficient data transmission, reliability and scalability. In industrial environments, the star topology is often favored, as in this structure each node is independently connected to a central hub via a physical cable. This structure basically corresponds to the controller-device model, in which the controller always has control over the star-shaped distributed devices. The fact that all data passes through the central node simplifies network management and troubleshooting.

When selecting a network topology, it is crucial to assess the available hardware resources, the patterns of application calls and the nature of the planned business processes, as these will influence the suitability of the topology for your specific requirements. In addition, the individual scalability requirements and the associated management effort should be taken into account.

Common topologies for industrial applications include not only star, but also mesh and hybrid configurations. The mesh topology, which is characterized by nodes that are connected to each other via multiple paths, offers high reliability and is suitable for networks in which communication must continue despite the failure of individual nodes. However, it is more complex and expensive to install due to the large number of connections required.

Hybrid topologies combine elements of different topologies in order to utilize their respective advantages and at the same time mitigate their disadvantages. For example, a hybrid network can use a star configuration at the core, while using a ring or bus layout for certain segments to optimize performance and fault tolerance.

It is also important to consider the impact of the chosen topology on network performance. The structure of the network can affect the speed of data transmission, latency times and the ability of the network to handle data traffic efficiently. In addition, the physical layout of the premises, the complexity of the network and the need for redundancy or fault tolerance are key factors in determining the most appropriate network topology.

To summarize, the selection of a network topology for industrial fieldbus or Ethernet networks should be based on a thorough assessment of the specific requirements of the industrial environment, including hardware resources, process types, scalability, management overhead and the desired level of reliability and fault tolerance.

With Hilscher hardware components for industrial communication at field level, flexible communication interfaces can be integrated into field devices and controllers. The netX SoCs, PC cards or embedded modules and their loadable firmware, i.e. the protocol stack software, can be integrated into any network topology that is supported by the respective communication protocol. For some solutions Hilscher also offers a topology editor to support users in setting up the correct topology for their specific use case.

Hilscher's netFIELD IIoT solutions can also be used regardless of the underlying network topology. The containerized netFIELD app PROFINET Tap, which analyses the data stream in PROFINET-controlled networks in real time and transmits process data from the devices to the MQTT broker via MQTT, automatically determines the PROFINET network topology during a network start-up phase and prepares it in its integrated web interface.