NETWORK TOPOLOGIES

The network topology is the link that ties end users to a data center and serves as the link between the devices within the data center. It is important to select a topology that fits the needs of the data center as well as the end users. If the data center is responsible for supporting a critical application and network uptime is of utmost importance, then a topology with multiple layers of redundancy, such as full-mesh, would be the best choice. This topology would help prevent network outages in the event that a cable or node in the network/data center fails. If the application supported by the data center is less critical and network outages would not cause a serious problem, then a less expensive topology, such as star or extended star, would be more appropriate.

Why is this relevant for C2G?

Our networking cabling products - copper and fiber - provide the links between the nodes within the topology.
​​​​​​​

Overview

A network topology describes how computer, printers, and other devices (i.e. nodes) are connected to the network. The following topologies are most commonly used to build most networks.
 

Bus Topology

---

A bus topology exists when all the nodes on the network are connected to a single cable. This single cable is commonly referred to as a backbone. Bus topology was used for early 10Base-2, ThinNet, and 10Base-5, ThickNet, coaxial cable Ethernet networks. In this topology messages sent from a node are broadcast to all nodes on the network. Only the intended recipient node accepts and processes the message. This type of network topology is relatively easy to install and inexpensive. This topology requires that both ends of the backbone cable be terminated. If the backbone is not terminated, then the signal is likely to bounce back from the end of the cable causing data collisions and noise that may disrupt the network. The main drawbacks to this type of network topology are a limitation on the number of computers that can be connected to the network, and the fact that only a single backbone cable is used to connect all of the nodes. Networks using a bus topology are limited to only a few dozen computers. If the network exceeds this size performance, problems will likely result. If there is a failure in the backbone cable connecting all of the nodes, then the entire network will become unstable and potentially cease to function. This topology is not typically used in modern networks.
 

Ring and Double Ring Topology

---

A ring topology exists when all of the nodes on the network are connected in a circle. Each node in the network acts as a repeater keeping the signal strong as it travels through the network. A node will generate a signal that is addressed to a specific computer on the network, and then the signal will be sent through the network in either a clockwise or counterclockwise direction. It is important to note that all signals on a network using this type of topology must travel in the same direction. This reduces the amount of data collision and noise on the network. The signal will continue through each node until it reaches the intended destination node. Typically, this type of network will use a Token Ring protocol, which allows only one computer to transmit a signal at any given time. The main drawback of this type of topology is that if there is a failure of any of the nodes or cables connecting the nodes, then the network will become unstable and potentially cease to function. The solution to this drawback is a double ring topology. The double ring adds a secondary cable for redundancy in the case of a failure. 
​​​​​​

Star and Extended Star Topologies

---

The star and extended star are the most popular topologies for Ethernet networks. This type of network is easy to setup, relatively inexpensive, and provides more redundancy than other topologies, i.e. bus topology. The star topology is configured by connecting all the nodes on the network to a central device. The central connection allows the network to continue functioning even if a single node or cable fails. The major drawback to this topology is that if the central device fails, then the network will become unstable or cease to function. The star topology is most suitable for small, centralized networks. The extended star topology adds sub-central devices that are connected to the central device. This type of topology is advantageous for large networks and provides functionality for the organization and subnetting of the IP address allocation within the network. The extended star topology is most suitable for large networks that may span an entire building.
 

Tree Hierarchical

---

The tree/hierarchical topology is configured by integrating multiple star topologies on a bus topology and using a central "root" node. The major drawback to this topology is that if the "root" node fails, then the network will become unstable or cease to function. This type of topology holds the advantage over a bus or star topology because it can better support future expansion of the network. However, this type of network is not commonly used because of the vulnerability of the topology.
 

Mesh Topology

---

​​​​​​​This topology is divided into two different types: full-mesh and partial mesh. A full mesh topology provides a connection from each node to every other node on the network. This provides a fully redundant network and is the most reliable of all networks. If any link or node in the network fails, then there will be another path that will allow network traffic to continue. The major drawback to this type of network is the expense and complexity required to configure this topology. This type of topology is only used in small networks with only a few nodes. A partial mesh topology provides alternate routes from each node to some of the other nodes on the network. This type of topology provides some redundancy and is commonly used in backbone environments, networks where services are vital, and in wide area networks, WANs. The most notable partial mesh network is the Internet.