Computer networks today use 2 layer models: the TCP/IP protocol model and the OSI (Open Systems Interconnection) reference model.
In this article you will learn what exactly the TCP/IP model is and where it actually came from, as well as learn about the set of existing protocols that fall under this model in the first and highest of its layers — the application layer. I will also give their characteristics and applications. If you want to learn more about IoT earlier, how information is transported between devices, and IoT applications, you can check out our article titled. “What is IoT (Internet of Things). How does it work and what are the types.”
Each application layer is a set of protocols that are used by applications to transport various data.
Let’s start with what the TCP/IP model actually is. Its name is derived from two names of communication protocols — Transmission Control Protocol (transport layer protocol) and Internet Protocol (network layer protocol). In the most general terms, TCP/IP can be described as a model for the layered structure of communication protocols (and so a collection of protocols), the origins of which can be traced to the 1970s. It was developed to create attack-resistant computer networks, but over time it became the basis for the Internet.
Interestingly, this model is often referred to as a protocol stack, due to the fact that the realization of the viewing of a given web page (and thus the communication of the user with the web server) requires the operation of several or even a dozen protocols. It is also worth mentioning that TCP/IP is much less detailed than the OSI model (mentioned at the beginning), as the second one describes very precisely what happens during network communication, but this detail is not always necessary). In the graphic below you will find the differences in the context of the number and naming of layers in both models, but we will focus on the OSI model in more detail another time.
The top layer of the TCP/IP model allows the user to use the network through software installed on the device. How? It mediates communication between programs and protocols of lower layers, thus enabling network interactions. In turn, each message sent thanks to the application, passes through all layers of TCP/IP. It is then transported to the other computer through the network, and then passes through all layers in the opposite direction. For the computer receiving the message, it looks like this: the message is sent from the network to the application layer and beyond.
It is worth mentioning that as data is transferred from the application to the network, each layer adds its own header to each message. These are then read by the corresponding layers on the computer receiving the information.
The application layer is an independent layer for other layer protocols and the transmission media used in them, so the application can be used whether the device is on a local, global or city network.
MAC header (Media Access Control) — usually refers to the Ethernet header, which is currently the dominant method of network access.
LLC header (Logical Link Control) — an interface to the MAC layer.
CRC MAC layer (Cyclic Redundancy Check) — the final field in an Ethernet MAC frame. The 32-bit CRC provides error detection when line errors (or Ethernet transmission collisions) cause MAC frame corruption.
Examples of protocols used at the application layer
- FTP (File Transfer Protocol)
Provides file transfer over the network.
- HTTP (Hypertext Transfer Protocol)
Responsible for transmitting web pages.
- HTTPS (Hypertext Transfer Protocol Secure)
An encrypted version of the HTTP protocol that uses TLS encryption.
- SMTP (Simple Mail Transfer Protocol)
To send e-mails.
- POP3 (Post Office Protocol v3)
Allows you to receive e-mails.
- IMAP (Internet Message Access Protocol)
POP3’s successor, is also used to receive mail, but additionally allows management of remote folders located on the server.
- DNS (Domain Name System)
Converts mnemonic names, i.e. domain names into corresponding IP addresses.
- TFTP (Trivial File Transfer Protocol)
A simplified version of FTP, used to install operating systems on network devices (such as routers).
Allows you to work in a text console.
- DHCP (Dynamic Host Configuration Protocol)
Assigns IP addresses, default gateway and DNS server addresses.
- SSH (Secure Shell)
Allows emulation of a network terminal providing encrypted connections.
- NFS (Network File System)
Used to share network drives.
- SMB (Server Message Block)
File sharing protocol.
- SNMP (Simple Network Management Protocol)
A protocol used to configure network devices and gather information about them.
In summary, in this article we started with a characterization of the TCP/IP protocol model, which is very important for further understanding of the layer descriptions. The application layer discussed first in the TCP/IP model is the highest of the four layers of the model mentioned, and can be said to be the “closest” layer to the network user. However, each layer plays its relevant roles, so you can certainly expect more posts on protocols on our blog — both in the OSI and TCP/IP models. Stay tuned!
Words by Kinga Kuśnierz, Content Writer