How does the Internet work
I bet that you use the Internet every day. And so do I! But how does the Internet actually
work? Read this guide and you'll understand what it's all about.
The Internet consists of a huge network of computers that are connected together. To
communicate properly, each computer must have a unique identifier, a unique IP
(Internet Protocol) address.
Network design
I design networks, and then I
bring them to life. Think
installation, implementation,
maintenance and full
documentation.
Performance
Improvements
I maximize network
performance by monitoring
performance, troubleshooting
network problems, scheduling
and implementing network
upgrades.
Network Security
I secure networks by
establishing, and then
enforcing security policies. I
install applications and
hardware that monitor and
prevent unauthorized access.
© Hondu Net. For Real!
My services
IP addresses consist of four (and lately six) groups of numbers and symbols that are
separated by dots. With the old IPv4 protocol, which is still largely used today, an IP
address would look like this:
123.44.32.93
Whenever you connect to your Internet service provider (ISP), your computer is assigned
a temporary IP address, which may change every time you disconnect from the ISP.
To communicate with the Internet, your computer must transform your requests (a search
phrase like "how to boil an egg", for example) into a data string that can travel to a
server that knows this answer, and is able to send it back to your computer.
A group of smart people have invented the TCP/IP protocol to do this. The protocol
includes several layers, which are specialized for accessing web pages, sending and
receiving emails, uploading and downloading files from FTP (file transfer protocol)
servers, and so on.
Of course, if your request utilizes too much information, it needs to be broken down into
smaller amounts of data, which have about 1,200 bytes each. These are called "data
packets", and their goal is to make it possible for the information to reach its destination
as quickly as possible, without being altered.
Each data packet is assigned a port number before being sent over the Internet. A
powerful domain name server (DNS) reads all these data packets, and then directs them
to the proper destination, according to their port numbers.
The DNS server is basically a computer which stores a huge database. Actually, things
aren't exactly that way, because the database is so big that it must be distributed
amongst a huge network of servers. But let's pretend that we're only talking about a
server and a database for now.
That server uses a huge table to determine the IP address of each computer that's
connected to the Internet. This is why whenever you type in (let's say) www.google.com,
you are taken to the home page of the biggest search engine. The DNS server has
translated this URL name, which makes sense to us, into a series of symbols that are
actually needed to reach Google's servers.
Once that your request has reached the destination, the target server can reply to it.
Surprisingly, the new data packets that are sent as an answer to your request don't
necessarily follow the same route. Why does this happen? Because it increases flexibility,
and our chances of getting the desired results as quickly as possible.
The open Internet architecture will help it survive even if a great number of node servers
go down. People may have to wait longer until the data packets reach their computers,
but the Internet will continue to work - isn't that fantastic?
Here's a practical example on how the Internet works. You are reading this article on your
monitor, but the actual article data is stored somewhere else. It may sound strange to
you, but this article is stored on my web server, where I've uploaded its text.
Before discovering my website, your computer has sent a request to your Internet service
provider, and then the data packets have traveled further, reaching a domain name
server. The DNS server has found a match for your request, leading you to my server's IP
address. The same thing happens anytime you send an email or chat with your friends
over Skype; you'll use a different layer of the TCP/IP protocol to do that, of course.
I'd like to praise the TCP/IP protocol a bit more, but frankly, I can't do that! It's an
unreliable protocol which has got lots of problems. And the sad thing is that it can't be
replaced, without us losing the ability of accessing many of our favorite Internet services.
But let's enjoy Internet as it is, for now. Who knows? Maybe someday people will find a
way of patching it without breaking it. How's that for ending the article in a positive note?
:)