There are more than 7,000 protocols globally, and their number is still increasing. Here we discuss the most frequently used rules for device interaction within networks.
A network protocol is a set of rules that governs how devices interact within a network. For successful information transmission and reception, all devices involved must adhere to and follow the protocol’s terms. Support for these protocols is integrated into either hardware or software components of the network or both.
The OSI or Open Systems Interconnection model facilitates communication between different protocols. Literally translated, OSI stands for “Open Systems Interconnection”.
OSI serves as a reference model for device interaction within networks.
The OSI model enables different communication systems to interact based on universally accepted standards. It’s akin to English being the Lingua Franca, the universal language in networking.
The model divides communication systems into seven distinct layers.
Should there be any disruptions in data transmission, the model aids in quickly and easily pinpointing and resolving issues at specific levels.
Although OSI remains a benchmark standard, it has become outdated as modern protocols now operate across multiple OSI layers. The OSI model has been superseded by the TCP/IP model used by most modern devices.
TCP/IP is the model that the Internet uses.
The TCP/IP model helps understand how nodes interact within the Internet network. Its name includes the two primary protocols that form the foundation of the Internet. TCP/IP stands for Transmission Control Protocol/Internet Protocol, encompassing both data transmission management and Internet protocols.
This model is prevalent across today’s Internet, with new network protocols being developed based on TCP/IP standards. For instance, when you connect to the Servercore website, you use IP, TCP, and HTTPS protocols that operate under this model.
We will next examine key protocols of network, transport, and application layers. We most frequently encounter these when analyzing any network or server issues.
Transport Layer Protocols: Brief Overview
Internet Protocol and IP addresses
The Internet Protocol (IP) is the simplest protocol that has connected individual computers into a global network. Its primary function is to route datagrams, determining the paths that the packets take across network nodes. Every device, such as your PC or printer, has an IP address to ensure data reaches the correct recipient. For instance, a file sent to print won’t accidentally be directed to your colleague’s personal computer instead of the printer.
A notable drawback of the protocol is its low reliability. It does not confirm packet reception and does not ensure data integrity. IP merely handles the sending process.
To send packets, it’s necessary to determine which port the packet should be sent to. To achieve this, the protocol utilizes its own addressing system. Addresses are either 32-bit (IPv4) or 128-bit (IPv6) formats. Before sending, a packet is provided with a header and payload (body of data to be delivered).
IPv4 is a 32-bit system divided into four sections (123.123.123.123). It supports up to 4,294,967,296 addresses and serves as the default protocol. Its primary advantage is simplicity. Its drawbacks include limited address space, also known as “address exhaustion”.
Conversely, IPv6 offers a 128-bit address space, providing roughly 2^128 possible addresses. The format consists of eight sections, each containing four hexadecimal digits. The protocol’s downside is the complexity it adds to network management. When renting a server or virtual machine from Servercore, IPv4 addresses are provided. However, you can request an IPv6 address as well. For VMware-based cloud systems, only IPv4 is available.
One of the key protocols operating over IP is TCP, often referred to as TCP/IP. However, it is not the sole protocol included in the Internet Protocol suite.
TCP facilitates message exchange across the Internet.
TCP aids network devices in exchanging messages. It functions at the fourth layer, the transport layer, of the OSI model.
For information transfer, the original file is segmented into parts that are sent to the recipient and then reassembled. For example, when a web page is requested, the server processes this request and responds with an HTML page via the HTTP protocol. It then requests the TCP layer to establish the necessary connection and dispatch the HTML file. TCP converts data into blocks and transfers them to the user’s TCP layer for transmission confirmation.
Properties of the TCP protocol:
- Segment Numbering System. TCP tracks each segment transmitted or received by assigning numbers to them. Data bytes set for transmission are given specific byte numbers, while segments receive sequence numbers.
- Flow Control. This function restricts the speed at which data is sent by the sender. This ensures reliable delivery of data. The receiver continuously informs the sender how much data can be accepted.
- Error Control. This feature enhances reliability by verifying the integrity of bytes.
Source Port and Destination Port. TCP uses special ports to facilitate communication among different protocols. For instance, SSH uses port 22, HTTP uses port 80, HTTPS uses port 443, and Gopher uses port 70. Ports are categorized into three ranges – well-known (0-1023), registered (1024-49151), and dynamic (49152-65535).
UDP – Counterpart to TCP: Outlining Differences in Network Behavior
Unlike TCP, the User Datagram Protocol transmits data without requiring user confirmation of the action’s result. This achieves greater speed and data transfer efficiency at the cost of reduced reliability and security.
The characteristics of the protocol dictate its specific applications. It is ideal for real-time applications like Skype and Discord where data transmission delays are critical. It’s also preferred for applications with many connected clients, such as in gaming, voice or video conferences, and streaming media.
UDP operates by assembling data into a UDP packet and appending its own header information to it. The UDP header comprises four fields, each 2 bytes in size: source port number, destination port number, header length, and block checksum.
UDP is favored by hackers for DDOS and DOS attacks. Since this protocol doesn’t require server confirmation, it allows attackers to overwhelm the server with requests. A typical attack involves sending numerous datagrams. This makes the server to respond to each one, consuming computational resources in the process.
SCTP – Stream Control Transmission Protocol
Another protocol that belongs to the transport layer. SCTP ensures reliable sequential data transmission. It supports multi-address connections where one or both endpoints may have more than one IP address. This allows to seamlessly switch between alternative network paths.
SCTP resembles UDP and TCP protocols that provide transport layer functions for certain Internet applications. As SCTP excels in rapid interface switching, only companies concerned with application downtime use it. SCTP operates over a non-contact packet network like IP and supports data transfers involving single or multiple IPs.
RTP – Real-time Transport Protocol
Real-time Transport Protocol is used for streaming audio and video, especially in IP telephony for voice transmission. RTP is used together with the RTCP control protocol for media broadcasting. While RTP broadcasts media, RTCP analyzes Quality of Service (QoS) statistics and synchronizes various streams. RTP uses even-numbered ports for transmission and reception, whereas RTCP utilizes odd-numbered ports.
The protocol is also regarded as the principal standard for transmitting audio and video over IP networks. RTP can simultaneously transmit to multiple end recipients using IP multicast.
Network Layer Protocols: Brief Overview
ICMP is a network control message protocol
Its role is to diagnose communication issues between devices. It determines whether the data has reached its destination or not.
The primary function of ICMP is to report errors. If any data fails to reach its destination, ICMP generates errors to communicate to the sending device. For instance, if the data volume is too large for the router, it will discard the packet and send an ICMP message back to the original data source.
Similar to UDP, ICMP can be exploited for network attacks like ICMP flooding and “ping of death”, where the key tactic involves generating a multitude of ICMP messages.
OSPF is a network link-state routing protocol.
Open Shortest Path First is used to determine the optimal route between the originating and destination routers. It operates at the network layer of the OSI model.
Once set up, OSPF will examine neighboring routers and gather all accessible link state information to construct a topological map of all viable paths within its network. It then stores this information in its topology database, also known as the Link State Database (LSDB).
Using the gathered data, it calculates the optimal shortest path to each available subnet or network via an algorithm known as Shortest Path First (SPF).
Application Layer Protocols: Brief Overview
FTP is a protocol for transferring data across a network.
FTP is a client-server protocol that utilizes two channels for data transmission: a command channel that manages the transfer process, and a transport channel that directly transmits the data. In FTP protocol, the end user’s device is referred to as the local host, while the second computer acts as the remote host, aka the server. The protocol requires correct configuration by the host and a specialized client on the local host.
Overview of Protocol Operations on the Internet
The user needs to log into the FTP server. It’s important to note that some servers allow access to part or all of their data without requiring authorization. This is referred to as “anonymous FTP”. In this case, files can only be transferred from the server to the client’s computer.
Subsequently, the client initiates a dialogue with the server, requesting permission to modify files on the server. By using an authorized FTP client, you can download files from the server, upload files to it, and perform other operations.
FTP sessions operate in two modes: active and passive.
- In active mode, the server opens a transport channel and begins transmitting data after initializing through the command channel.
- In passive mode, the server uses the command channel to send the client the necessary data to open a data transmission channel.
Since the client establishes all connections in passive mode, this protocol is particularly effective for use with firewalls.
DNS – directory of Internet domain names
Browsers interact through IP addresses. To connect to a website, people use its domain name, such as https://servercore.com/. Domain Name System converts domain names into IP addresses, enabling the loading of internet resources via a browser. Each device on the network is assigned a unique IP address, which other devices use to connect to it, while the DNS server saves people from having to remember these addresses.
Currently, there are four main DNS servers involved in the web page loading process:
- The DNS Recursor acts as a directory, handling requests from user computers like browser applications.
- The root nameserver is the initial step in converting a hostname into an IP address and provides a list of DNS servers.
- TLD nameserver is the next step in finding an IP; it stores information about all domain names sharing a common extension (.ru, .com, etc.).
- Authoritative nameserver provides final responses to DNS inquiries.
HTTP(S) – Hypertext Transfer Protocol
HTTP is a cornerstone of the Internet, facilitating web page loading via hypertext links. It belongs to application protocols and operates on top of other network protocol stack layers.
Typically, HTTP data transfer involves a client computer sending a request to a server, which then responds. Each HTTP request contains encoded data with various information, including:
- HTTP version,
- URL,
- HTTP request method – indicating an expected request from the server (e.g., PUT and GET requests),
- header – which transmits basic request information and includes key-value pairs,
- request body (optional, includes any transmitted information).
Upon receiving a request, the server is required to respond. Its standard structure comprises a status code, a header, and a response body.
HTTP request status codes are three-digit numbers that typically reflect whether the request was successful. These are divided into five primary categories:
- 1xx* Informational,
- 2xx Success,
- 3xx Redirection,
- 4xx Client Error,
- 5xx Server Error,
*XX represents numbers from 00 to 99.
Like a request, a response includes a header containing various information, such as the language of the transmitted data. It usually contains HTML data that the client’s web browser converts into a webpage.
Discussing HTTP necessitates mentioning its lack of security. During data transmission, everything occurs openly, allowing for potential interception by malicious actors. To address this issue, the HTTPS protocol was developed. A detailed comparison of these protocols can be found on our blog.
SSH is the primary method for connecting to servers
SSH, or Secure Shell, is a secure protocol primarily used for server connections. It allows users to log into an existing account on the server where all commands sent are executed.
This connection follows a “client-server” model. A program known as a daemon must be operational on the remote device to establish this connection. The daemon connects to a specific network port, authenticates connection requests, and sets up an appropriate environment if the correct credentials are provided. Appropriate software must also be installed on the client side.
In the Servercore control panel, users can upload their own SSH key for enhanced connection security or generate one on-site.
Outdated protocols include Telnet, Gopher, and FTP
Today, there are over 7 million different protocols in existence. As the network evolves, certain protocols become obsolete, such as Gopher, FTP, and Telnet. Let’s examine the latter in more detail.
Telnet is an ancient yet highly reliable communication protocol. Originally developed for character-oriented terminal emulation within UNIX systems. Telnet is extensively used today for managing routers, switches, remote servers, and for basic text communication that doesn’t require graphics.
Conclusion
This article explores popular protocols for interacting between devices and software across networks. In reality, hundreds of protocols are used by just your PC alone, each designed for specific tasks. Details about a protocol’s tasks and operating principles can be found in a document known as an RFC Standard.
