www.howtonetwork.org

CCNA & CCNP Certifcation Training

  • About
  • Contact
  • FAQ
  • Join
  • Home
  • CCNA
    • ICND1
    • ICND2
    • 101 CCNA LABS
  • CCNP
    • ROUTE
    • SWITCH
    • TSHOOT
    • 101 CCNP LABS
  • CompTIA
    • Security+
    • Network+
  • Design
    • CCDA
    • CCDP
You are here: Home / ICND1 / Day 7 – IPv6 / Day 7 Tasks / History of IPv6

History of IPv6

Fit for Purpose?

When Sir Tim Berners-Lee devised the World Wide Web in 1989, there was no way he could have predicted the huge impact it was to have on the world. Personal computers were prohibitively expensive and there was no easy way to communicate over long distances unless you could afford expensive WAN connections. Even then, there was no agreed communication model for all to follow.

Something needed to change and change came in the form of a new addressing standard for IP. Learning from mistakes made and responding to changes in business requirements, the Internet Engineering Task Force (IETF) published the first of many IPv6 standards as far back as 1998.

There will be no switch-over date; instead, networks will gradually transition to running both IPv4 and IPv6, and then eventually IPv4 will be phased out of existence. At the moment, approximately 1% of all Internet traffic is running on IPv6 (source: Yves Poppe, IPv6 – A 2012 Report Card).

Why Migrate?

I’ve already said that when IPv4 was devised, the Internet wasn’t used by the general public, and why would they? There were no websites, no e-commerce, no mobile networks, and no social media. Even if you could afford a PC, there wasn’t much you could do with it. Now, of course, almost everybody is online. We carry out most of our day-to-day tasks using the Internet, and businesses rely on it to exist. Soon we will be using mobile devices to manage our cars and home security, to turn the coffee maker on, to set the heating level, and to set the TV to record our favourite show.

Some of this is already taking place, not only in Europe and the Americas but also in fast-developing countries such as India and China where billions of people live. IPv4 simply isn’t up to the job and even if it was, there aren ’t enough addresses to cater for demand.

Here are a few benefits to changing to IPv6:

  • The simplified IPv6 Packet header
  • Larger address space
  • IPv6 addressing hierarchy
  • IPv6 extensibility
  • IPv6 broadcast elimination
  • Stateless autoconfiguration
  • Integrated mobility
  • Integrated enhanced security

I’d like to delve into packet layer analysis of IPv6, as well as the many types of headers available, but there isn’t space here to do so, and since it isn’t tested in the exam there is no need to include it. Instead, I will focus on what you need to know for the exam and your role as a Cisco engineer.

Hex Numbering

It may be well worthwhile to have a short memory jogger on hex numbering.

You know that decimal numbers consist of 10 digits ranging from 0 to 9. Binary consists of two digits ranging from 0 to 1. Hex numbering ranges from 0 to F and has 16 digits. These addresses are also referred to as base 10, base 2, and base 16, respectively.

You can see that each numbering system starts with a zero, so:

Decimal – 0,1,2,3,4,5,6,7,8,9

Binary – 0,1

Hex – 0,1,2,3,4,5,6,7,8,9,A,B,C,D,E,F

When you write these addresses, you may not realise it but you are using columns from right to left; the rightmost is the one column and the next column is the base number times the preceding column, so:

Numbering Base

N to 3rd power

N to 2nd power

N to 1st power

N

10 – Decimal

1000

100

10

1

2 – Binary

8

4

2

1

16 – Hex

4096

256

16

1

You can see that each successive column from the right increases in value. For decimal numbering it is 10 multiplied by 1. For binary it is 1 and then 1 multiplied by the numbering system of 2. If you compare the three numbering systems up to the last hex digit, you can begin to see why hex is the preferred format for IPv6 addressing.

Decimal

Binary

Hex

0

0000

0

1

0001

1

2

0010

2

3

0011

3

4

0100

4

5

0101

5

6

0110

6

7

0111

7

8

1000

8

9

1001

9

10

1010

A

11

1011

B

12

1100

C

13

1101

D

14

1110

E

15

1111

F

In order to provide enough addresses for our needs many years into the future, IPv6 has been designed to provide many trillions of available addresses. In order to do this, the numbering range has been expanded from 32 binary bits to 128 bits. Every 4 bits can be represented as one hex digit (as can be seen from the chart above). Logic then dictates that two hex digits will give us 8 bits, which is a single byte, or octet.

An IPv6 address is 128 bits in length and this is broken down into eight sets of 16 bits each separated by a colon when written in full format. Every 4 hex bits can range from 0000 to FFFF, with F being the highest digit available in hex numbering:

0000

0000

0000

0000

0000

0000

0000

0000

to

to

to

to

to

to

to

to

FFFF

FFFF

FFFF

FFFF

FFFF

FFFF

FFFF

FFFF

Next

About Us

This is a free bonus site for members of www.howtonetwork.com

Copyright

The content on this copyright Reality Press Ltd.
Copyright Reality Press Ltd.