Understanding the basics of network switches

Networking is a big topic. In fact, it’s too big for a single conversation. In order to distill such a massive concept into manageable bites, it’s best to look at one component of a network at a time. This time around, that bite is network switches. They come in many shapes and sizes, but you can navigate through the field of options pretty easily when you break switch features into three categories.

Managed vs Unmanaged

These two options have a simple trade-off. Unmanaged switches are much easier to deploy. They come with pre-configured networking protocols, security and traffic management. They are typically plug-and-play devices that can make simple and fast additions to an existing network.

Managed switches are the opposite. They require manual configuration before they can be fully deployed. That component grants vastly superior control over security, resource distribution and the overall functionality of the switch. The time and effort spent on deploying managed switches returns a more secure, better managed network (provided the network administrator does the job correctly).

Choosing between these configurations is not a matter of want. They serve separate roles in a robust network. Major nodes should be run by managed switches. They can do the heavy lifting. Where a network might need rapid changes, unmanaged switches shine. Since they are ultimately tied into the properly managed nodes anyway, the easy deployment makes them perfect for simple additions to the whole.

Ports

The point of a switch is to add ports to a network. When choosing, you have to consider three points. The first that comes to mind is speed. Everyone wants a fast network, and in enterprise networks, that want is more of a need. If a switch cannot match the bandwidth capacity of the greater network, then it creates a bottleneck. That isn’t always bad, but you absolutely need to consider the speed of the switch. It can range from 10/100 (fast Ethernet) to tens of Gigabits per second.

Your second consideration is interface. Fast networks often have fiber optics carrying the workload, but fiber optic cables don’t come with a single connector type (unlike Ethernet). In a large enough network, there’s a good chance you’ll be working with a variety of connectors (ST, SC and LC just for starters), and each switch needs to support all of the interfaces expected of it. Double check your connectors before any purchase.

The third consideration only applies to Ethernet, but it’s widely important since most of your devices will tie into the network via Ethernet or WiFi. In particular, Power over Ethernet (PoE) adds a lot of versatility to any network switch. As the name implies, it can carry power for a device through the Ethernet line. This is ideal for endpoint devices that don’t thrive on battery power. Security cameras and WiFi repeaters are two devices that come to mind.

Housing

Switches have to be stored somewhere, and the housing design is going to have a lot of say in where you can put it. Mostly, the two varieties are stackable and standalone. Standalone switches are best when they are physically isolated. If you have a single switch working as a relay for a few devices, standalone is the way to go. Stackable are designed for networking closets and server-style housing. They can be organized more efficiently than standalones, and they can be configured in bulk. They’re a requirement for major nodes on a large network.

This covers the basics of switches. There are still plenty of additional considerations. We haven’t touched on routers, firewalls, signal repeaters, cables or any of the other big topics, but with this overview, you should have a good feel for what you want out of your switches. When you know the ports you need, the housing style and how a switch will be managed, the rest is just trying to get the most performance and features for you buck.

To order or if you have any questions, please contact one of our Network Equipment Experts today.