There are two types of networking environments: traditional networking environments and controller-based networking environments. In this tutorial, we will compare both and understand the main differences between them.
Traditional networking environments
In traditional networking environments, we individually configure and manage each network device. For example, if we have 10 routers in our network, we need to configure and manage them separately.
Configuring and managing each network device is a tedious task. Since we configure each device manually by entering commands, there is also a possibility of mistakes.
With traditional networking environments, we take the following actions for each network device over the lifetime of the device:
- We install the network device (such as a router or a switch) into the network.
- We make the initial configuration of the device. The initial configuration allows the device to perform its functions in the network.
- We monitor the overall performance of the device and events that occur on the device. For this, we can use network management protocols such as SNMP (Simple Network Management Protocol) and NetFlow.
- Over time, we modify and update the configuration as per the requirements of the network.
- We upgrade the operating system on the device as needed.
At first glance, the above-listed tasks may not seem like a lot of work. But if we have to repeat these tasks with each device on the network, it becomes an enormous task.
For example, if we want to update the IOS, we have to take the following steps.
- Backup the running configuration and old IOS
- Add the new IOS and configure the device to use the new IOS to boot
- Restore the running configuration from the backup
To update the IOS on all routers, we have to perform the above-listed tasks 30 (3 steps x 10 routers) times.
To make the situation more complex, for each configuration change we want to make, we need to log into the router. Logging into the router is not as easy as logging into the PC. To log in to the router, we need to access the router through a console connection or remote protocols like Telnet and SSH. After making changes, we must have to disconnect the device. We have to repeat this process on each device whenever we want to make changes.
Key points:-
- Administrators manage all devices independently. To manage them, they have to establish a new connection each time. They also have to terminate the connection after making the changes.
- Administrators manually make all changes. Manual configuration is prone to configuration errors.
- It uses the blacklisting security model. In this model, everything is allowed unless you disallow it.
Controller-based networking environments
Controller-based networking uses automation technologies to apply configuration changes to multiple devices at the same time. It uses the controller software or device that automates configuration changes on the network and manages the deployment of configuration changes to multiple devices simultaneously.
It actively monitors the network for changes. When we add a new device to the network, it automatically applies the initial configuration to the device, without needing our intervention.
We can use automation for many tasks. For example, we can use it to manage IOS images on devices. Once configured, it actively checks the developer's site for new versions. If the new version is available, it automatically takes the backup of the existing version on the device and replaces it with the new version. All this process is done automatically without our intervention. We also don't need to visit the devices.
Another use of automation is task scheduling. We can schedule specific tasks or operations on the device and automate troubleshooting procedures if there is a problem with the device.
Key points: -
- Administrators manage all devices collectively. To manage them, they use the controller software or appliance.
- Administrators only need to configure the controller software or appliance. The controller software or appliance manages all devices automatically based on their configuration.
- Since the controller configures all devices, the chances of configuration errors are zero unless the administrator configures the controller device incorrectly.
- It uses the whitelisting security model. In this model, everything is blocked unless you allow it. It enhances the network's security.
- It reduces network costs. Since the maximum work is done by the controller software or appliance and a company needs only one or two administrators to manage the controllers, it does not need to hire many administrators.
- It saves time. With traditional networking, administrators need to manually configure all devices which takes too much time.
- It provides consistency in configurations. Administrators create the script and then instruct the controller to deploy that to the devices. This way all devices receive the same configuration.
