TCP/IP Basics

Internet communication does not run on one large mechanism alone.

When you view a website, send email, or an app connects to a server, multiple communication rules work in combination.

The foundational idea behind this is TCP/IP.

The name TCP/IP includes TCP and IP, but in practice it does not refer only to TCP and IP. It is used as a term for the group of communication protocols used on the internet, in other words the overall basic mechanisms of communication.

This article organizes TCP/IP not as a detailed specification, but as an entry point for understanding the overall picture of internet communication.

What TCP/IP Is

TCP/IP is a basic protocol suite used for communication on the internet.

A protocol is a set of rules for communication. Rules are needed for things such as how to specify a destination, what units to use when sending data, how to handle data that does not arrive, and what format applications use to exchange data with each other.

In TCP/IP, communication is considered by separating it into roles.

In this way, internet communication is established by combining IP, TCP, and UDP as foundations with surrounding mechanisms such as DNS, HTTP, and TLS.

IP Directs Communication Toward the Destination

IP stands for Internet Protocol.

The central role of IP is to use IP addresses to direct communication toward the destination. On the internet, IP addresses are used to identify devices and servers.

For example, when a browser communicates with a web server, packets are ultimately sent toward the destination IP address. Routers along the way look at information such as the destination IP address and decide where to send the packet next.

However, IP is a mechanism for "directing toward the destination," and by itself it does not guarantee the reliability of communication as a whole.

IP alone cannot sufficiently handle things such as the following.

• Whether the data arrived
• Whether the data arrived in the correct order
• Whether lost data should be resent
• Which app inside one device it should be passed to

TCP, UDP, port numbers, and higher-level protocols are involved in these matters.

TCP Emphasizes Reliability

TCP stands for Transmission Control Protocol.

TCP is a mechanism that makes it easier to handle data accurately. It establishes a connection with the other endpoint, confirms whether sent data arrived, resends data if any was lost, and arranges data that arrived out of order into the correct order.

For this reason, TCP is suitable for communication where missing data or changed order would cause problems.

However, TCP is not a mechanism that makes everything fast. Because it performs delivery checks, retransmission, and similar control, latency may increase depending on communication conditions.

TCP becomes important in situations where reliability matters more than low latency.

UDP Emphasizes Lightness and Flexibility

UDP stands for User Datagram Protocol.

UDP does not perform connection establishment, delivery checks, order control, or retransmission in the protocol itself as TCP does. Because of that, the mechanism is lightweight and it is easier to design necessary control on the application side.

UDP is used for communication where real-time behavior or flexible control is important.

UDP does not mean "bad communication because it is unreliable." UDP itself provides fewer guarantees and leaves necessary control to the application or higher-level mechanisms.

There are also mechanisms such as QUIC that use UDP as a foundation while implementing their own reliability and encryption.

Port Numbers Distinguish Applications

An IP address indicates the network location of the other endpoint. However, multiple applications and services communicate at the same time inside a single device or server.

For that reason, an IP address alone does not tell us which application the communication should be passed to.

This is where port numbers are used.

For example, when web, DNS, email, and other services run on the same server, they can be distinguished by different port numbers.

Even if TCP and UDP use the same number, they are handled as separate communication. In other words, TCP port 443 and UDP port 443 are not the same communication even though they use the same number, "443."

DNS, HTTP, and TLS Also Run on TCP/IP

When you view a website, IP, TCP, and UDP alone do not complete the whole process.

People usually specify websites by domain name, not IP address. For that reason, DNS is used first to look up the IP address corresponding to the domain name.

Once the IP address is known, the browser communicates with the destination. HTTP is used to retrieve web pages. For HTTPS, HTTP communication is protected by TLS.

The important point here is that these mechanisms do not compete with each other; they divide roles.

For example, HTTPS is communication where HTTP is protected by TLS. And in many cases, TCP or UDP and IP are used underneath it.

Flow When Opening a Web Page

From the perspective of TCP/IP, the flow of opening a web page can be organized simply as follows.

In actual communication, caches, CDNs, load balancers, and reuse of existing connections are also involved. For that reason, every step is not always performed from the beginning in exactly the same order every time.

However, as a basic model, it becomes easier to organize the process if you think of DNS, IP, TCP/UDP, TLS, and HTTP as having separate roles.

TCP/IP and Anonymity

TCP/IP is not a mechanism for anonymization. It is the foundation for making communication work on the internet.

For that reason, information needed to establish communication may be visible on the communication route or to the destination.

Using HTTPS or TLS makes communication content easier to protect. However, IP addresses, traffic volume, communication time, relationships with destinations, and similar information do not all disappear.

When thinking about anonymity, it is necessary to distinguish between "communication content" and "outside information needed to establish the communication."

Summary

TCP/IP is the basic protocol suite for establishing internet communication.

IP uses IP addresses to direct communication toward the destination. TCP establishes a connection and emphasizes reliability through delivery checks, retransmission, and order control. UDP provides fewer guarantees and emphasizes low overhead and flexibility.

Port numbers are used to distinguish which application or service inside a single device or server the communication should be passed to.

DNS, HTTP, TLS, and similar mechanisms also work by dividing roles on top of TCP/IP. When opening a web page, DNS looks up the IP address, IP directs communication toward the destination, TCP or UDP handles how data is delivered, TLS protects communication for HTTPS, and HTTP exchanges page content.

TCP/IP is not anonymization technology. To establish communication, information such as IP addresses, port numbers, communication protocols, traffic volume, and time is handled.

Understanding this foundation makes it easier to organize "which layer and which mechanism" is being discussed when learning about DNS, HTTP/HTTPS, TLS, VPNs, Tor, communication logs, and similar topics.

Related tools

Related articles