Basics of networking

The intercommunication of devices in a computer network consists of several stages, with several components. Each of them is equally important, as each performs the tasks required for proper communication. These steps are defined by the so-called hierarchical model. Anyone familiar with the layered model knows that an understanding of this is the basis for further knowledge and skills in the field of computer networks.

There are two layered models, the TCP/IP protocol model and the ISO/OSI reference model. On the one hand, they are similar to each other, and on the other hand, each model communicates slightly differently. However, before we discuss these two models and explain the differences between them, we will tell you
you why and when you should use them, what they are used for and what the benefits of using them are.

Dividing the network communication process into layers brings many benefits, the most important of which are:

·         easier definition of communication rules and principles (these are communication protocols),

·         the ability to work with network equipment and software from different manufacturers,

·         it is easier to understand the possibility of the whole communication process,

·         ability to manage the communication process.

Before the data from the source device reaches the end device, it has to travel a long way, during which it is first appropriately labelled, tagged, described with specific information allowing it to be identified, and then transferred between a number of intermediary devices until it reaches the recipient, who must then translate it.

Without such a model, which divides communication into smaller, more comprehensible and manageable stages and defines the tasks that need to be carried out in each layer, it would be difficult to manage network communication properly, because the numerous solutions and technologies create a huge chaos, uncontrolled. Imagine a situation where there is no such build-up, no rules describing communication, and each hardware and software manufacturer creates its own independent system.

Of course, in one company's solution, communication will be very efficient and fast, but the solutions of two separate companies may be incompatible with each other. In practice, we use network hardware and software from different companies, thanks to the division into separate layers with rules and tasks describing their operation. These rules and tasks are the same for everyone, but each company, each manufacturer, be it hardware or software, can implement them in its own way.

A typical example is operating systems. Some users use Windows, some come from a Linux distribution and some come from macOS. Each of these systems is different and each performs web tasks in a different way, but ultimately
on each of these systems, a web page or email will look the same or at least similar. Therefore, some of the most important benefits of using the hierarchical model include:

·         management of the network communication process,

·         defined rules and tasks,

·         interoperability at the hardware and software level between network products from different manufacturers,

·         and control of the correctness of communication.

Now that we know the purpose of the hierarchical models, let us move on to discuss their most important features. Both models originated a long time ago in the 1970s, but are still current and in use today. The first is the TCP/IP model, known as the protocol model. Each of its layers performs specific tasks using specific protocols. On the other hand, the ISO/OSI models, known as reference models, are more commonly used for analysis in order to better understand the communication processes taking place in a network and are models for the design of network solutions, both hardware and software.

In the case of the TCP/IP model, we can distinguish 4 layers, these are Application, Transport, Internet and Network Access.

The application layer allows users to use web services such as the web, email, file sharing, terminal connections and instant messaging. This is the layer closest to the user because it allows us to take full advantage of the benefits of modern web services. For example, when we sit in front of a computer and launch a web browser, we are using the web at the application layer level.

Below this is the transport layer, whose main task is to handle communication between devices efficiently. At this layer, the data is split into smaller parts,
and then supplemented with additional information, allowing them to be distributed to the appropriate application on the target device and mounted on the target device in the correct order.

Then there is the internet layer, whose main task is to find the shortest
and fastest route to the target device over the WAN, but also uses logical addresses (IP addresses) to address the data.

Finally, we have the network access layer, which encodes the data as pure bits (zeros and ones) and passes them to the transmission medium and addresses them, this time via a physical address (MAC address).

The ISO/OSI model consists of 7 layers (application, presentation, session, transport, network, data link, physical).

At the top end of this model, we can distinguish the application layer, which works here very similarly to the TCP/IP model in that it enables network applications to be used by network end users.

Then there is the presentation layer, which transmits information to the application layer about the data format used, e.g. informs which file types will be transmitted, and is responsible for the correct encoding of the data on the source device and decoding on the target device.

Below this is the session layer, which manages user sessions via website or video communication, for example.

Going one step further, we have the transport layer, which again is exactly the same as in the TCP/IP model and in both cases the function of this layer is exactly the same.

Then there is the network layer, which is the equivalent of the internet layer of the TCP/IP model, i.e. has very similar functions such as addressing and determining the best path to transmit data.

Next we have the data link layer, whose main task is to control access to the transmission medium and address the data, but this time to transport it between hosts on the LAN.

Finally, the physical layer encodes the data into pure bits (1s and 0s) and transmits them over the transmission medium to the appropriate device.

The two models are very similar. The resulting difference can be seen in the upper layers; in the case of the ISO/OSI model, it is divided into 3 layers, whereas
in the case of the TCP/IP model, the same function is performed by only one layer. The layers can be seen with a similar difference, in the ISO/OSI model we have two separate data link layers and physical layers, whereas in the case of the TCP/IP model there is only one network access layer.