Connected Magazine

Main Menu

  • News
  • Products
    • Audio
    • Collaboration
    • Control
    • Digital Signage
    • Education
    • IoT
    • Networking
    • Software
    • Video
  • Reviews
  • Sponsored
  • Integrate
    • Integrate 2024
    • Integrate 2023
    • Integrate 2022
    • Integrate 2021

logo

Connected Magazine

  • News
  • Products
    • Audio
    • Collaboration
    • Control
    • Digital Signage
    • Education
    • IoT
    • Networking
    • Software
    • Video
  • Reviews
  • Sponsored
  • Integrate
    • Integrate 2024
    • Integrate 2023
    • Integrate 2022
    • Integrate 2021
FeaturesNetworking
Home›Features›IPV6 – what is it and where is it? – Part I

IPV6 – what is it and where is it? – Part I

By Geoff Meads
14/02/2022
0
0

Hot property tends to move fast and IP addresses have been all the rage for quite a while. Geoff Meads explores the evolution of IPV6 and limitless possibilities.

Go back 17 years or so and you’ll find the announcement of a new version of the IP (Internet Protocol) addressing scheme. For clarity, an IP address operates at layer three of the Object Systems Interconnection (OSI) model and identifies a network host within the context of the network it is currently connected to.

Today we have tens of billions of devices connected to the Internet, all working (mostly) just fine. So why do we need something new?

ADVERTISEMENT

The answer is in the maximum number of possible IP addresses. In short, we’ve run out!

IPv4 vs. IPv6 – The Numbers

When the current IPv4 format was first conceived (approximately 1981) the Internet we have today could barely have been imagined. At that point the ‘new’ IPv4 scheme promised 4,294,967,296 addresses or a little under 4.3 billion. Back then, the thought that one day there might be more connected devices than that would have been thought ludicrous. Fast forward 15 years or so and that number was already looking woefully inadequate.

So, in 2006 the IPv6 scheme was launched and the number of possible IP addresses grew beyond all comprehension.

The actual number of possible IP addresses in an addressing scheme is derived from the number of binary bits each address contains. Let’s look at a typical IPv4 address:

192.168.100.150

Here we see four decimal numbers, each separated by a dot. Each decimal number can have a value from 0 to 255. This is because each decimal number block within an IPv4 address (an ‘Octet’) is made up of 8 binary bits. Adding this up we have 4 x 8 = 32 binary bits (or 232 addresses).

Now let’s look at an IPv6 address in comparison:

2001:0db8:85a3:0000:0000:8a2e:0370:7334

Even at first glance we can see a couple of clear differences between an IPv6 address and it’s IPv4 predecessor:

1.       It’s much longer

Rather than a maximum of 12 characters in the decimal IPv4 notation we now have 32 characters in the IPv6 notation.

2.     It’s Expressed in Hexadecimal

Where the binary bits of an IPv4 address are written as their decimal equivalent number the bits of an IPv6 address are written as hexadecimal numbers. This means that each character can represent 16 possible values (0-9 then a-f) rather than the ten values (0-9) of a decimal number.

Extracting this to binary, an IPv6 address has 128 binary bits rather than the 32 of an IPv4 address. Or, put another way, 2128 addresses.

So how many IPv6 addresses are there? Well, it’s a big number. A really big number:

340,282,366,920,938,463,463,374,607,431,768,211,456 to be exact or approx. 3.4×1038

The IoT Connection

One of the key concepts of IPv6 is that all devices can be directly connected (with appropriate security) to all other devices. This is not the case with IPv4.

Currently, using IPv4, we have two types of network,  a Local Area Network (or LAN) like we have in our home and a Wide Area Network (the Internet). While the separation of these with a router and firewall afford some security between LAN devices and the rest of the world, the actual reason these exist is to help us reuse IP addresses.

One of the fundamentals of connected networks is that the scope of IP addresses in directly connected networks cannot be the same. For example, you can’t have the address 192.168.1.1 as a valid address on two networks that are directly connected. If you did, and a packet was bound for that address, a router between the two networks wouldn’t know which network to send the packet to.

In practice millions of LANs around the world use the same IP address scope (e.g. 192.1.8.1.0 / 24). This is possible because they are never directly connected, the WAN sits between them and that has a different IP address scope.

Moving to IPv6, there will no longer be a need for separate LAN and WAN networks as far as addressing is concerned and every device in the world can have its own IP address. The reuse of IP addresses in multiple networks won’t be necessary.

This concept was initially known as the ‘Internet of Things’ (IoT) and was connected to the introduction of IPv6. However, as IPv6 take-up has been slow and the number of connected devices using IPv4 rose anyway, the term IoT came into general use, even without IPv6.

Shortening Addresses

One of the downsides of IPv6 addresses is their length. They are both long and not exactly memorable! With this in mind a number of shortening strategies are allowed to make the job of using and documenting IPv6 addresses easier.

Here are two key examples:

Zero Values

Where there are four zeros together (in one or more consecutive blocks), we can remove them. In fact, as long as the numbers continue to be zeros, you can remove all of them:

2001:0db8:0000:0000:1a2c:8a2e:0000:0334 can be shortened to: 2001:0db8::1a2c:8a2e:0000:7334

Any equipment interpreting the address will assume the missing characters are all zero. Note that You can only do this once! For example, you cannot remove the second block of zeros further into the address. This is NOT ALLOWED: 2001:0db8::1a2c:8a2e::0334

In addition any block of four zeros can be shortened to a single zero, therefore this is ALLOWED: 2001:0db8::1a2c:8a2e:0:0334

Leading Zeros

We can also remove leading zeros so this: 2001:0db8::1a2c:8a2e:0:0334 becomes: 2001:db8::1a2c:8a2e:0:334

Subnets

Just like IPv4 addresses IPv6 addresses contain two parts, a network ID and a host ID. Let’s take a look at an example IPv4 address ‘192.168.10.123 / 16’.

If you are familiar with IPv4 you’ll see this is network with a 16 bit subnet mask i.e. the first 16 bits (‘192.168’) are the network ID and the second 16 bits (’10.123’) are used as the host ID.

Exactly the same is true for IPv6 addresses. Taking our original (un-shortened) address of 2001:0db8:0000:0000:1a2c:8a2e:0000:0334, and assuming a 64 bit mask (‘/64’), we see that the network ID is ‘2001:0db8:0000:0000’ while the host ID is ‘1a2c:8a2e:0000:0334’

And That’s Just the Beginning

Hopefully you’re already seeing that, while IPv6 brings a huge increase in scope and some added complexity in documentation, there are many aspects of the new scheme that are identical to that in IPv4.

Next time we’ll look at how and where IPv6 is being used and where we might expect it to go next.

  • ADVERTISEMENT

  • ADVERTISEMENT

Previous Article

A WALL-SMART fit for Sonos soundbars

Next Article

CEDIA APAC AWARDS 2021 – Integrated Home ...

  • ADVERTISEMENT

  • ADVERTISEMENT

Advertisement

Sign up to our newsletter

Advertisement

Advertisement

Advertisement

Advertisement

  • HOME
  • ABOUT CONNECTED
  • DOWNLOAD MEDIA KIT
  • CONTRIBUTE
  • CONTACT US