When you open a browser and go to a webpage, data is traveling through the internet, to a server to get the content. It then travels back to serve you that content on your computer. But how does data travel on the internet, really? I mean, how does it go to a server and back that fast?
In networking, data is called packets. These packets are what matters and they have everything needed in them, like the destination address and the request. To understand it better, let’s dive deeper into how data actually travels over the internet in this article.
A data packet, or network packet to be correct, is a collection name of several components. These components are needed in order for you to be able to browse the web, for example. In order for you to go to a website, you will need to send packets to the websites web server and that server will have to send packets back to you. For this to work, a packet needs to hold several components:
First, the packet needs two addresses, the source address, and the destination address. How would it else know where to go and where it came from?
A packet has something called “time to live”. This is the number of hops the packet can do until it dies. Each time the packet goes through a network node (i.e a router or a switch, like a crossing) the number is decreased. This is to avoid packets with errors hopping around the web for eternity.
Depending on what the packet carries, one packet may not be enough. Each packet consists of a length so that the receiver knows if the packet came alone or if more is to come.
This is the actual data, the content of the packet.
A packet can also include other components, but these are the main ones that are always there. Network packets are how information (data) is being carried over the internet. But how does the packet travel?
As you know, this data packet will not just magically appear somewhere else, it needs to find the destination and then get there. For the package to know where to go, something called routing will be used. This is essentially just your local post office carrying out post. The mailman looks at the zip code, then the address, then the house number and then the name, and data packets do the same thing.
How the data is being transmitted is through cables. It might start off as a package on your mobile phone and will wirelessly be transferred to your router. WiFi signals are working like waves, much like radio signals. I highly suggest taking a look at this article from vice.com, where they have imagined how WiFi would like if we could see it.
Once the router has received the signals from your phone, the packet will continue through cables, from your router and out to the world wide web. Your house is most likely connected to the grid in your city. The packet has its address written on it and will go where the signs are pointing.
The signs, in this case, is called DNS, or Domain Name Servers. These exist to tell other services, servers, clients and data packets where to go. Often, a DNS will point to another DNS that is closer to the destination. However, if the DNS knows where the destination is, it will point directly there instead.
As you can tell, it is very much like a parcel, motorways, and crossings. It’s very easy to see the similarities and that is because it works in the same way. Why change something if it works?
Fast. No, really, it’s fast.
In a fiberoptic cable, which is what companies are digging down, the speed for a data packet is about 200 000 km/s (124 300 miles/second. To give you some sort of reference on how fast that is, the circumference of the Earth is about 40 000 km. This means that the packet can go 5 times around the globe in a fiberoptic cable.
But we are not done. Because packets can move even faster in the air. Without interference, they can reach speeds of up to 99.7% the speed of light, or 299 100 km/s. That is two hundred thousand one hundred kilometers per second or 185 723 miles per second. That is insane. In the air, the packet can go 7 times around the Earth, at the same time it takes you to blink your eye.
Now, to be real, it’s not possible to count as that all the time as there are obstacles in the way. Things like routers, switches, and other nodes do not support this kind of speeds (unsurprisingly) so they are the bottlenecks. Think about that the next time you are thinking that your internet is slow.
This is the good thing with the internet, and something that has made the internet become so big as it actually is (what I think anyway). There is no one that owns the internet. There is no government, no company, no organization. There is no CEO on the top that can go bad and shut it down any minute.
Since the internet is a hard thing to grasp, let’s focus on what it relies on instead: hardware. Here, many organization and companies provide their hardware for the internet to work efficiently. Your ISP, for example, has servers, switches, and routers that make their part of the internet working as it should. They are then connected to another company or organization that take over.
So, if all companies in the whole world press the off-button at the same time, the internet would go down. On the other hand, that would never happen as these companies make tons of money on the internet, and money talks 🙂
Because of the above, there is no one that is responsible for the overall internet. There are organizations that will standardize technology and infrastructure, so that is easier for these companies to talk to each others. You may have heard of some of these organizations, like ICANN (Internet Corporation for Assigned Names and Numbers) or IETF (Internet Engineering Task Force). However, these organizations are not responsible for the internet.