How Computer Networks work

---

The OSI (Open System Interconnection) model was developed by the International Standards Organization to lay a framework for how open systems should communicate. There is nothing here you can see or touch, but it is just for people who work in the industry to get an understanding of how data is passed from device to device and processed. Understanding this model can help in systems development and in troubleshooting networking issues. There are seven layers to the OSI model.

7 Application

6 Presentation

5 Session

4 Transport

3 Network

2 Data Link

1 Physical

The Application layer is the layer in which the human interfaces with the computer system. Logging in, creating and saving files as well as file transfers all take place at this level. Email applications are also a part of this layer.

The Presentation layer transforms data into the form that the application accepts. This layer formats data to be sent across a network. It is sometimes called the syntax layer.This layer works to present data between applications in a format in which the application accepts the data.

The Session Layer establishes, manages and terminates connections between Applications.

The Transport Layer provides for the transparent transfer of data between the source and the destination hosts.

The Network Layer is the layer provides the functionality that connects data from one node to another on different networks. This is where routing takes place. Network routers work at this level. Some (Layer 3 ) Switches also work at this level. 

The Data Link layer connects two or more nodes on a network and also provide error correction. MAC (Medium Access Control)  and LLC (Logical Link Control) both work at this level. MAC is responsible for controlling how devices in a network gain access to a medium and permission to transmit data.LLC is responsible for identifying and encapsulating network layer protocols, and controls error checking and frame synchronization.

The Physical Layer is the layer that cabling and hubs work at. This is your Physical end to end connections without error correction. This is just the basic electrical, physical hardware connections.

 Now that we have an idea of what computer networks look like, let's take a closer look at the hardware components of the computer network.

Network Interface Card - NIC:  

A network Interface card is the device that connect the end device (Computer, Laptop, tablet or smartphone) to a network. Tablets and Smartphones usually have internal integrated wireless network interface cards that you cannot see. Laptops also have internal integrated wireless cards, but can also have wired network cards that you can plug a network cable into. Desktop Computers or Personal computers and servers usually have a wired network card in them. Wired and wireless network cards both have their advantages and disadvantages. Wired networks require you to run wire to each location where you want to put a computer. This can sometimes mean running wires through walls and ceilings. This can be time consuming and expensive, but a wired connection is extremely reliable. Wireless networks on the other hand can be easier to deploy, but can lack the reliability of a wired network. Network cards are also known as Ethernet NICs. Ethernet defines the type of standard hub or switch used in local area networks. Wireless Ethernet is also known as Wi-Fi.

Below on the left you will see a photo of a USB wireless ethernet or (Wi-Fi) adapter. On the right is a standard ethernet adapater that is connected by an ethernet patch cable.

Bridge:

A bridge can be used to connect two different Local Area Networks together. This usually works at the Layer 2 level of the OSI model. It uses MAC addresses like a layer two switch to decide where to send the data packets. 

 Router:

A router is a device that connect 2 or more networks together. It creates a routing table to know where to send data packets destined for a particular network. This works at the layer 3 level of the OSI model. Routers can even connect to networks that use different types of data connections. It can connect an Ethernet LAN to a Broadband connection over the internet. 

Modem:

Modem stands for Modulator Demodulator. It is a network device that translates a carrier signal to a digital signal your computer can understand. In the past modems were used to send data over voice lines. This was very slow. With the invention of DSL we can now send digital signals over those same lines at much higher speeds using DSL modems.

Wireless Router:

A Wireless Ethernet router is a device that allows the user to create a local area network using wireless network cards and then connect to the internet. The router can accept both wireless and wired connections and then connect to a DSL or broadband cable network. 

Wireless Access Point:

A wireless access point looks like a wireless router but usually can only connect wireless devices to the local area network. This does not mean that you cannot connect to the wireless access point and not have internet access. It just means that the wireless access point does not connect directly to the internet but must forward your packets to another router that connects you to the internet. These are usually used in medium to large organizations that need wireless access points spread over a large area to allow many devices in many locations access to the network over a wireless or Wi-Fi connection. 

Firewall:

A firewall is a hardware device or software that controls access between the local area network and the internet. Some routers also have firewall capabilities. Firewalls usually have a little more advanced technology to help filter traffic and inspect data packets than a standard router. A Firewall can have much more control over traffic than a traditional router. They are used to stop unwanted traffic from crossing over your network. A firewall such as the one pictured earlier can also break networks up into Virtual Lans like a switch and then give those networks each a number based on how much that network is trusted. The internal network would be the trusted network, the external Internet would be untrusted and then web and email gateway servers would be on a medium trusted network called a DMZ (Demilitarized Zone). The DMZ is where you would put your servers and devices that allow connections from the Internet. This protects your internal network by keeping that untrusted traffic on a different network.

Network Cabling

Ethernet cables - Cat 5, Cat 5E, Cat 6 and Cat 6 A: 

Ethernet cable is made up of twisted pairs of copper wire covered by shielding. In side the cable are 4 pairs of wire twisted together which is why ethernet cable is also known as twisted pair. There are various classes of cabling based on the data speed you need it to handle. Ethernet cable can only be run in lengths of around 100 meters or about 328 feet before it needs to be terminated or connected to another device. You can plug the cable into a switch or a hub and then run another cable for another 100 meters. This is important to know when you are planning your network. Ethernet cabling like we stated earlier comes in different classes known as categories or cat for short. Cat 5 cabling can handle network speeds of up to 100 megabits per second. Cat 5 E (the e means Enhanced) can handle up to 1 gigabit per second. Cat 6 and Cat 6 A can handle up to 10 gigabits per second so it is ten times faster than Cat 5 E. Cat 6 A (the A is for augmented) is designed to handle interference better than standard Cat 6. The reason that we need to keep the cable length to 100 meters or less is because of something called Attenuation. As the data moves over the wire the signal degrades after a certain distance. In order to maintain optimal performance you want to make sure you do not run your ethernet cables longer than the recommended distance. Also be careful to keep your wires away from other devices or electronics that may create interference which can also degrade your signal. 

Fiber Optic Cabling:

Fiber optic cabling is designed to carry beams of light at high speed over long distances. It is much more expensive than ethernet. Like ethernet there are different grades of cable for different uses. Two types of fiber cables are called single mode and multi-mode fiber. Single mode fiber uses very thin glass strands and a laser to generate light while multi-mode fibers use LEDs. Different fiber cables are rated at 10 gigabits per second, 40 gigabits per second and 100 gigabits per second.Data can also travel much farther over a fiber optic cable than ethernet because light does not degrade the way and electrical signal does. Some fiber can be run up to 80 or 100 kilometers. That is about 50 to 60 miles. On each end of the fiber cable is a fiber transceiver that takes the electrical signal and converts it to light pulses and when data it received it take the light pulses and converts it back to an electrical signal. 

When buying and installing fiber optic cabling it is important to check with your vendor that you get cable and transceivers with the correct specifications so that they work together.

.