You are currently viewing Network Basics

Network Basics

In this modern era where everything is connected, it is important to understand how networks work. A lot of devices nowadays have a form of network connection. Sometimes even without knowing this. 

Keep in mind that we are in the middle of a huge transition. In the past, physical networks formed the biggest part of a network. Small Office/Home Office (SOHO) and On-Premise networks were the bread and butter of any network specialist. But nowadays we have been moving into the cloud and with COVID-19 the transition to the cloud has been accelerated. The only issue with this is that many networks still have a physical (on-premise) component which means that local system admins and network hobbyists enter an era of Hybrid Cloud: partially cloud and partially on-premise. Because of this, an understanding of both Cloud Networks and On-Premise Networks is important.

I already dedicated a few posts to Virtual Machines (a very important part of Cloud Networks) but as said before, this is only a very small part of the vast world of networks.

The role of hosts, clients, and servers in a network

A host is a computing device that is connected to a network. A local host is a device that you are using: your workstation, your telephone, Internet of Things devices, etc. 

In addition to a local host, there are also remote hosts. Remote hosts are other computing devices on the network or reachable beyond the network. 

Each network host has a specific role. Let’s take the internet as an example. A remote computer that is called a “Web server” stores the files that create a Website. This means that the Web server has the role to provide access to Websites. Two popular Web server programs are Apache HTTP Server and Microsoft Internet Information Services (IIS).

Your Web browser (Chrome, Edge, Firefox, etc.) sends a request to the Web server to share Web page files. After that, the files are displayed on your device. Because your device asks for a web page, it is called the client. This is the role of the local host in this example. A remote computer that serves the Web site is the server. Any computer that is running a program that is shared with clients (local hosts) is a server by definition. 

Another example is a small home network. Let’s say that you have two computers in your home and that one computer is connected to a printer with a USB wire. This computer has enabled a printer-sharing program that is built into Windows. Because of this, other computers on the network can use the printer as well. Because of this, the computer that is attached to the printer has the role of print server because it serves the printer.

Networks are always used to share and access all kinds of things, called resources in general between computing devices. The only limitation in sharing and accessing resources over a network in your own ability to find a server program that is capable of sharing it and a client program that can access it. 

In order to be able to share and access resources, a network needs to have the following prerequisites:

  • Something that defines and standardizes the design and operation of cabling, network cards, and the interconnection of multiple computers: a Network Interface Controller (1)
  • A method of sharing resources and accessing those shared resources: Ethernet (2).
  • A method to address clients in order to enable them to find servers and the ability to enable servers to send data to those clients, no matter the size of the network: a Local Area Network (3).

Network Interface Controller (1)

In order to be able to communicate, clients and servers need a Network Interface Controller (NIC). Network Interface Controllers define or label the machine on a network. A NIC also breaks files into smaller units of data to send across the network and reassembles the units it receives into complete files. 

When the NIC breaks down the files into smaller units, these units are called frames. NICs create and process frames. You might have heard the word packet instead of frames when people are talking about sending data over wire or WiFi. That is incorrect. Packets are included in a frame. 

NICs have a built-in identifier: a unique address of the network card. This address is called a Media Access Control (MAC) address. If you would like to know the MAC address of your device and if you are using Windows, go to the command-line interface (enter “CMD” in the Windows search bar) and enter “ipconfig /all”: . This way you can see the address of your NIC (MAC address). It is named “Physical Address”. ‘

Ethernet (2)

In order to share resources, you need infrastructure. Ethernet was initially developed for wired networking but also wireless networks use Ethernet as the basis for their signals. 

Ethernet was invented in the mid-1970s by a consortium of companies that included Digital Equipment Corporation, Intel, and Xerox. This consortium wrote a series of standards that defined everything that was necessary to transfer data from one computer to another. They called this series of standards Ethernet. Most Ethernet networks employ one of three speeds: 10BaseT (10 Mbps), 100BaseT (100 Mbps, called “Fast Ethernet”), or 1000BaseT (1000Mbps, called “Giga Ethernet”). 

Ethernet uses Star Bus topology. With this technology, all individual hosts are connected to a central box. Each system is then attached to this box with cables to special ports. The box sorts the frames that are received from all individual hosts and sends them to the hosts where the send frame needs to move to. The “bus”  refers to the internal wiring in the box and the “star” refers to the wires that lead from the box to the hosts. The magical box in the middle is called the switch. A switch provides a common point of connection for network devices. Most switches have 4 to 8 ports but bigger networks in organizations use 32 or more ports. In the early days of Ethernet, networks used a hub but hubs are long gone and nowadays only switches exist. The difference between a hub and a switch is that a hub was sending signals to all connected ports. Because of this unnecessary traffic, hubbed networks were very slow. Switches memorize the MAC addresses of all the connected devices and only send out signals to the correct host. This is why switched networks are much faster than hubbed networks: each system in a switched network uses full bandwidth and this was not the case with hubs. This is also the reason why hubs are obsolete nowadays.

The cabling that is used between the switch and the system is called Unshielded Twisted Pair (UTP) cabling. UTP has different types of categories, called “CAT”. Nowadays the most common form of CAT cabling is CAT 6. CAT 6 cabling supports 1000BaseT networks at 100-meter segments and 10-Gbps networks up to 55-segments.

In addition to UTP cabling, you also have Shielded Twisted Pair (STP) cabling. STP has a shield that covers the wires in the cable in order to protect them from Electromagnetic Interference (EMI). STP is only used in locations with excessive electronic noise, like shop floor areas with lots of lights, electric motors, or other machinery that can cause problems for UTP cables.

At the end of the cable, there is a connector: the RJ-45 connector. This is the standard for UTP connectors and has connections for up to four pairs.

Local Area Network (3)

A Local Area Network (LAN) is a group of computers that are physically located and close to each other: no more than a few hundred meters apart. The group of computers in a LAN “hear” each other when one of them sends a broadcast. A group of computers that is connected by one or more switches is called a broadcast domain. All nodes (computers) in this domain receive broadcast frames from every other node. Setting up a LAN in a Small Office/Home Office is a bit different compared to setting up a basic LAN in a corporate environment.

SOHO

If you use a LAN in a Small Office/Home Office (SOHO), the most common way is to use wireless technology (WiFi). Another option is using the electrical network in the building for connectivity. This option is called Ethernet over Power. This requires a specialized bridge that connects to power outlets. In case you have a computer in a place where wireless internet doesn’t work and where UTP cables can’t reach, you can use Ethernet over Power.

Ethernet over Power Bridge

Corporate

An Ethernet LAN in its basic configuration has a switch, some UTP cables, and a few computers. This is called a “Star Network”. 

Star Network

Setting up a Star Network is less basic than the above image. It is possible to put all cables directly to a switch in the middle but this would mean that people had to cross those cables. This ends in tripping over cables and damaging cables. 

Because of this, a successful setup of a basic Star Network requires three essential ingredients: a telecommunications room, horizontal cabling, and a work area. Below picture shows what this looks like:

Basic Star Network Setup in an office

Final Thoughts

I hope this post provides more insight into how a network works and how it is set up. As stated before, we are entering an era where physical networks on-premise are starting to disappear and where Cloud computing gets the upper hand. However, it is good to know the basics of a network because Cloud computing still requires a part of these basics.

Feel free to contact me if you have any questions or if you have any additional advice/tips about this subject. If you want to keep in the loop if I upload a new post, don’t forget to subscribe to receive a notification by email.

Leave a Reply