IP, TCP and Computer Stacks

Here's a piece of ComSci homework we had to complete about protocols, networks and the Internet.

IP, TCP and Computer Stacks

By Minghua Yin, 10W

What is a typical Internet packet made up of?

An Internet packet contains control information and user data (also known as the payload). The control information is usually found in the header and trailer of the packet, with the payload in between. An IP packet, for example, contains an IP header and a payload. The header, a bit like in emails and letters, contains data about the packet itself, including the type of IP packet (IPv4 or IPv6), the IP address of the destination machine and the length of the packet. The payload is like the “message” and contains data such as part of the content of a website.

How does an Internet packet move through the Internet?

The Internet is essentially a network of routers all sending and receiving bytes of information. An IP packet may be sent by a local router and travel through the network to a router that covers a wider area. The router then does a longest-prefix lookup on the IP address of the packet’s destination, where the bits of the IP address are compared with a table of IP addresses to determine which way – through which interface – the packet should travel.

What is a server?

A server is a computer or computer program which manages access to a centralized resource or service in a network.

What is a client?

A client is a desktop computer or workstation that is capable of obtaining information and applications from a server, or a program that is capable of obtaining a service provided by another program.

What is a dynamic IP address?

A dynamic IP address is one that changes every time a user logs on to his or her computer. The IP address is assigned each time by the user’s ISP (Internet Service Provider) using the DHCP (Dynamic Host Configuration Protocol) protocol.

Why are IP addresses dynamic?

IP addresses are usually dynamic for security reasons. Dynamic IP addresses are also more cost effective than static ones, and there is no human intervention, so there is less risk of a conflict or network problem. Static IP addresses are better for VOIP (Voice Over Internet Protocol), online gaming, game hosting and VPN (Virtual Private Network), as there is less chance of the service disconnecting, since your IP address is always the same; however, the benefits of dynamic addresses usually outweigh the negatives.

What is the difference between UDP and TCP?

The UDP (User Datagram Protocol) and TCP (Transmission Control Protocol) protocols both work at the transport layer of the OSI model, but have different uses. TCP is more reliable than UDP – it is connection-oriented, and so guarantees that the packets sent are all received in the correct order, whereas UDP is connectionless, so some data may be lost. This means that UDP is more lightweight than TCP, as in TCP resend requests may have to be sent if a packet goes missing, and the packets received need to be ordered if not already, which takes time; UDP, on the other hand, is “fire and forget”! In TCP, the data is read as a stream, with no indication as to when one packet ends and another begins, and there may be multiple packets per read call, whereas in UDP, packets are sent individually.

Therefore, TCP is better for the World Wide Web, emails and files, where all the data needs to be received completely, while UDP is better for DNS, media streaming, VOIP and online games, where complete data retrieval is not necessary and the retrieval needs to be fast.

Extra stuff

A day in the life of a packet (sent from a business computer)…

First, the data that needs to be sent is packaged and control information (headers and trailers) added. The packet then travels through the LAN via router switches, maybe passes into a broader intranet via a router, and eventually reaches a proxy, where it is blocked if it is requesting banned data as defined by the proxy, such as a blocked website. Otherwise, the packet continues on its journey and reaches the firewall, where it is blocked if it is carrying sensitive corporate information. If it makes it through the firewall, it is transferred by the Internet router onto a bandwidth, which connects to the Internet. When it arrives at its destination after travelling across the Internet, there may be a firewall, which prevents viruses and other malware from getting through, as well as packets requesting to go through a closed port. The packet travels to its destination machine, such as a server or another computer, maybe making a short journey through the intranet and LAN first. Finally, the packet is unpacked and the data it is carrying read.

PackagedàLAN (router switches and routers)àProxyàFirewallàInternet routeràBandwidthàFirewallàPortàInterfaceàUnpackaged

Difference between the intranet and LAN

A LAN (Local Area Connection) connects the router with your computer and other personal devices, usually in close or direct contact with one another. An intranet is like an expanded LAN, connecting many different devices in a certain area, such as devices in different offices at a business.