Introduction to Networks

Week 8 — IPv6 Subnetting

Reza Farashahi — NZSE

Quick Review

Week 6 — IPv6 Fundamentals Recap

IPv6 Address — Quick Recap

128 bits long (vs. 32 bits in IPv4)
Written as 8 groups of 4 hex digits separated by colons
- Example: 2001:0DB8:0000:0000:0000:0000:0000:0001
Simplification rules:
- Drop leading zeros in each group: 0DB8 → DB8
- Replace the longest run of all-zero groups with :: (once only)
- Simplified: 2001:DB8::1

Global Routing Prefix
48 bits (assigned by ISP)

Subnet ID
16 bits

Interface ID
64 bits (host part)

IPv6 Address Types

Type	Prefix	Meaning	Example
Global Unicast (GUA)	`2000::/3`	First 3 bits = `001`. Routable on the internet — like a public IPv4 address.	`2001:0DB8:ACAD:1::1`
Link-Local (LLA)	`FE80::/10`	First 10 bits = `1111 1110 10`. Only valid on the local link; auto-assigned on every IPv6 interface.	`FE80::1`
Unique Local (ULA)	`FC00::/7`	First 7 bits = `1111 110`. Covers `FC00::–FDFF::`; in practice only FD00::/8 is used. Like private IPv4 (10.x, 192.168.x). Not routable on the internet.	`FD12:3456:789A:1::1`
Loopback	`::1/128`	Equivalent to `127.0.0.1` in IPv4. Used to test the local TCP/IP stack.	`::1`
Multicast	`FF00::/8`	First 8 bits = `1111 1111`. One-to-many delivery. Replaces broadcast — no broadcast in IPv6!	`FF02::1` (all nodes)

Reading the prefix notation: FC00::/7 means the first 7 bits of the address are fixed (1111 110). The 8th bit can be 0 or 1, so the range is FC00:: through FDFF::. The IETF currently only defines the FD range (8th bit = 1), so in practice Unique Local addresses always start with FD.

Multicast — who uses it and why? A multicast address is a destination-only address — it is never assigned as a source. When a device sends to a multicast address, every device that has joined that multicast group receives the packet. Common groups:
• FF02::1 — All nodes on the local link (every IPv6 device listens on this — replaces broadcast)
• FF02::2 — All routers on the local link (only routers listen on this)

Review Quiz

Q1: How many bits are in an IPv6 address?

bits

Q2: What does the :: in an IPv6 address represent?

The subnet mask One or more groups of consecutive all-zero blocks The default gateway

Q3: Match each address to its type. Select the correct type from the dropdown:

Address	Type
`FE80::1`
`2001:DB8:ACAD::1`
`FF02::1`
`FD00:ABCD:1234::1`
`::1`

Review Quiz (continued)

Q4: True or False — “IPv6 uses broadcast to send packets to all devices on a network.”

True False

Q5: What prefix do Unique Local addresses start with in practice?

Q6: What is “dual stack”?

Running two switches together Running IPv4 and IPv6 on the same device simultaneously Using two IP addresses for one interface

IPv6 Simplification Practice

Q1: Simplify: FE80:0000:0000:0000:0210:A4FF:FE01:0023

Q2: Simplify: 2001:0DB8:0000:00A0:0000:0000:0000:0001

Q3: Simplify: FF02:0000:0000:0000:0000:0000:0000:0001

Q4: Expand the shortened address back to full form (8 groups, all leading zeros):

2001:DB8::FF:1

Rule	Before	After
1. Remove leading zeros Drop leading 0s in each 16-bit group	2001:0DB8:0001:00A0	2001:DB8:1:A0
2. Keep at least one digit `0000` becomes `0`, not empty	0000	0
3. Replace longest run of all-zero groups with `::` Must be consecutive `0` groups	2001:DB8:0:0:0:0:0:1	2001:DB8::0:1
4. Use `::` only once! If two equal-length runs exist, replace the leftmost	2001:0:0:1:0:0:0:1	2001::1:0:0:0:1

IPv6 Subnetting

Dividing an IPv6 Network into Smaller Subnets

IPv4 vs IPv6 Subnetting

IPv4 Subnetting

Borrow bits from the host portion
Worry about how many hosts you need
Complex binary maths (block sizes, broadcast addresses…)
Example: 192.168.1.0/24 → /26 = 4 subnets

IPv6 Subnetting

Uses the Subnet ID field (bits 49–64)
Host part is always 64 bits — never touched!
Much simpler — just change the 4th hextet
Example: 2001:DB8:A::/48 → /64 = 65,536 subnets

💡 Key difference: In IPv6 subnetting, we never borrow from the 64-bit Interface ID. The host part always stays /64. This makes IPv6 subnetting much easier than IPv4!

IPv6 Address Structure

A typical IPv6 Global Unicast Address has three parts:

Global Routing Prefix
48 bits
Assigned by ISP / registry

Subnet ID
16 bits
You control this!

Interface ID
64 bits
Host address — never changes

Written as 8 groups (hextets):

            2001:0DB8:AAAA :
            0001 :
            0000:0000:0000:0001
          

↑ Prefix (48 bits) ↑ Subnet (16 bits) ↑ Interface ID (64 bits)

The /48 Allocation

Organisations typically receive a /48 prefix from their ISP
This means the first 48 bits are fixed (the Global Routing Prefix)
You get 16 bits for the Subnet ID (bits 49–64)
2¹⁶ = 65,536 subnets available!

48 bits — Prefix (fixed) 16 bits — Subnet ID 64 bits — Interface ID

💡 Bottom line: With a /48, you can create up to 65,536 individual /64 subnets — each supporting an astronomical number of hosts (2⁶⁴ ≈ 18 quintillion).

How IPv6 Subnetting Works

It’s simple — just change the 4th hextet (the Subnet ID):

Subnet	Network Address	Prefix Length
0	2001:DB8:AAAA:0000::/64	/64
1	2001:DB8:AAAA:0001::/64	/64
2	2001:DB8:AAAA:0002::/64	/64
3	2001:DB8:AAAA:0003::/64	/64
…	…	/64
FFFF	2001:DB8:AAAA:FFFF::/64	/64

That’s it! The prefix stays the same, the 4th hextet counts up (0000 → 0001 → 0002 … FFFF), and the Interface ID is always available for hosts.

Hexadecimal — Quick Refresher

IPv6 uses hexadecimal (base 16). Here’s how it maps to decimal:

Hex	0	1	2	3	4	5	6	7	8	9	A	B	C	D	E	F
Dec	0	1	2	3	4	5	6	7	8	9	10	11	12	13	14	15

Each hex digit = 4 bits
Each hextet (4 hex digits) = 16 bits
After F (15), we carry over: F + 1 = 10 (hex) = 16 (decimal)
After FF (255), next is 100 (hex) = 256 (decimal)

💡 Count in hex: 0, 1, 2, … 9, A, B, C, D, E, F, 10, 11, … 1F, 20, … FF, 100 …

Hex ↔ Binary Practice

Part A — Convert Hex to Binary (4 bits each)

Hex	Your Binary
`A`
`F`
`3`
`C`

Part B — Convert Binary to Hex

Binary	Your Hex
`1110`
`0101`
`1001`
`1011`

Bonus — Convert full hextet to binary

Convert the hextet CAFE to binary (16 bits):

Example 1

Subnetting a /48 into /64 Subnets

Example 1 — The Scenario

Your company has been assigned the IPv6 prefix:

          2001:DB8:CAFE::/48
        

You need to create subnets for 4 departments:

Engineering
Sales
Finance
IT Support

Question: What will the network address be for each department’s subnet?

Step 1 — Identify the Parts

With a /48 prefix:

48 bits — Global Prefix (fixed) 16 bits — Subnet ID (your choice) 64 bits — Interface ID

The prefix 2001:DB8:CAFE takes up the first 48 bits (3 hextets)
The 4th hextet is the Subnet ID — this is what we change
Each subnet will be a /64
We just need to assign a different value in the 4th hextet for each department

Step 2 — Assign Subnet IDs

Simply count up from 1 in the 4th hextet:

Department	Subnet ID	Network Address
Engineering	0001	2001:DB8:CAFE:1::/64
Sales	0002	2001:DB8:CAFE:2::/64
Finance	0003	2001:DB8:CAFE:3::/64
IT Support	0004	2001:DB8:CAFE:4::/64

💡 Notice: The prefix stays the same (2001:DB8:CAFE). Only the 4th hextet changes. Each subnet has room for 2⁶⁴ hosts — more than enough!

Step 3 — Assign Router Addresses

By convention, the router gets ::1 as the Interface ID on each subnet:

Subnet	Router Interface Address
Engineering	2001:DB8:CAFE:1::1/64
Sales	2001:DB8:CAFE:2::1/64
Finance	2001:DB8:CAFE:3::1/64
IT Support	2001:DB8:CAFE:4::1/64

And PCs in each subnet can use any Interface ID, for example ::10, ::11, ::A1, etc.

Comparison to IPv4: This is just like using .1 for the router gateway in IPv4 subnetting (e.g. 10.0.0.1).

Your Turn — Example 1 Practice

Given the prefix 2001:DB8:CAFE::/48, what would be the network address for a 5th subnet (e.g. “Marketing”)?

What would be the router address on that 5th subnet?

With a /48, how many total /64 subnets can you create?

Example 2

Subnetting with a /32 Prefix

Example 2 — Larger Allocation

An ISP gives you a /32 prefix:

          2001:DB8::/32
        

32 bits — Prefix 32 bits — Subnet ID 64 bits — Interface ID

Subnet ID = 64 − 32 = 32 bits
Number of /64 subnets = 2³² = 4,294,967,296 (≈ 4.3 billion!)
Hextets 3 and 4 are both available for subnetting

Example 2 — Creating Subnets

With a /32, the 3rd and 4th hextets form the Subnet ID:

Subnet	Network Address	Usage
1	2001:DB8:0001:0000::/64	Auckland Office
2	2001:DB8:0001:0001::/64	Auckland WiFi
3	2001:DB8:0002:0000::/64	Wellington Office
4	2001:DB8:0002:0001::/64	Wellington WiFi

💡 Strategy: Many organisations use the 3rd hextet for site/location and the 4th hextet for VLAN/subnet within that location. This creates a clean, hierarchical scheme.

The Simple Formula

To find how many /64 subnets you can create:

Number of /64 subnets = 2^{(64 − prefix length)}

Prefix You Receive	Subnet Bits	/64 Subnets Available
/32	64 − 32 = 32	2³² = 4,294,967,296
/36	64 − 36 = 28	2²⁸ = 268,435,456
/40	64 − 40 = 24	2²⁴ = 16,777,216
/44	64 − 44 = 20	2²⁰ = 1,048,576
/48	64 − 48 = 16	2¹⁶ = 65,536
/56	64 − 56 = 8	2⁸ = 256
/60	64 − 60 = 4	2⁴ = 16
/64	64 − 64 = 0	2⁰ = 1 (no subnetting)

Practice — How Many Subnets?

Q1: You receive the prefix 2001:DB8:1234::/48. How many /64 subnets can you create?

Q2: You receive the prefix 2001:DB8:AB00::/40. How many /64 subnets?

Q3: You receive the prefix 2001:DB8:FACE:FF00::/56. How many /64 subnets?

Example 3

School Network — Step by Step Subnetting

Example 3 — School Network

NZSE has received the following IPv6 prefix from their ISP:

          2001:DB8:ACAD::/48
        

They need subnets for:

Subnet	Purpose
1	Staff Network
2	Student Labs
3	Student WiFi
4	Servers
5	Security Cameras
6	Guest WiFi

Example 3 — Solution

Subnet	Purpose	Network Address	Router (::1)
1	Staff	2001:DB8:ACAD:1::/64	2001:DB8:ACAD:1::1
2	Labs	2001:DB8:ACAD:2::/64	2001:DB8:ACAD:2::1
3	Student WiFi	2001:DB8:ACAD:3::/64	2001:DB8:ACAD:3::1
4	Servers	2001:DB8:ACAD:4::/64	2001:DB8:ACAD:4::1
5	Cameras	2001:DB8:ACAD:5::/64	2001:DB8:ACAD:5::1
6	Guest WiFi	2001:DB8:ACAD:6::/64	2001:DB8:ACAD:6::1

💡 This is the same prefix we used in Practical 6! The subnets 1, 2, 3 match the VLANs 10, 20, 30 from your lab.

Your Turn — NZSE Network

NZSE wants to add a 7th subnet for IoT Devices. What is the network address?

A PC on the Staff network (Subnet 1) has Interface ID ::A1. What is its full IPv6 address?

How many total subnets could NZSE create with their /48 prefix?

Nibble Boundaries

In IPv6, a nibble = 4 bits = 1 hex digit. Subnetting is easiest when the prefix falls on a nibble boundary (a multiple of 4):

Prefix	On Nibble Boundary?	Easy to Read?
/32	✅ Yes (32 ÷ 4 = 8)	Very easy
/48	✅ Yes (48 ÷ 4 = 12)	Very easy
/52	❌ No (52 ÷ 4 = 13.0)	Trickier
/56	✅ Yes (56 ÷ 4 = 14)	Easy
/60	✅ Yes (60 ÷ 4 = 15)	Easy
/64	✅ Yes (64 ÷ 4 = 16)	Very easy

💡 Good news: In practice, ISPs almost always assign prefixes on nibble boundaries (/32, /48, /56), so IPv6 subnetting stays straightforward.

Example 4

Home Network with a /56 Prefix

Example 4 — Home Network

Your home ISP gives you a /56 prefix:

      2001:DB8:1234:AB00::/56
    

Q1 — Subnet bits: How many bits are available for subnetting?

Hint: Subnet bits = 64 − Prefix length

Q2 — Number of subnets: How many /64 subnets can you create?

Hint: 2^{subnet bits}

Q3 — Subnet range: The Subnet ID occupies the last 2 hex digits of the 4th hextet. What is the range of values it can take?

From to

Your Turn — /56 Practice

Given 2001:DB8:1234:AB00::/56, what is the network address of the 10th subnet?

Hint: Decimal 10 = Hex 0A

What is the last subnet you can create with this /56? (Subnet 256)

Common Mistakes to Avoid

❌ Don’t Do This

Borrowing from the Interface ID
- In IPv6, the last 64 bits are always for the host
Using broadcast addresses
- IPv6 has no broadcast — use multicast instead
Trying to calculate “usable hosts”
- With 2⁶⁴ hosts per subnet, this is irrelevant

✅ Do This Instead

Always subnet within the Subnet ID field
- Between the prefix and bit 64
Use the formula: 2^{(64 − prefix)}
Keep subnets at /64
- This is standard practice for all IPv6 networks
Use ::1 for router gateways
- Convention, just like .1 in IPv4

IPv4 vs IPv6 Subnetting — Side by Side

Aspect	IPv4	IPv6
Where to subnet	Borrow from host bits	Use Subnet ID field only
Host part size	Varies (depends on mask)	Always 64 bits
Calculation	Binary maths + block sizes	Just change the 4th hextet
Broadcast address	Must calculate for each subnet	Doesn’t exist
Usable hosts	2ⁿ − 2 (important!)	2⁶⁴ (irrelevant)
Typical example	172.16.0.0/16 → /19 = 8 subnets	2001:DB8::/48 → /64 = 65,536 subnets
Difficulty	Complex	Simple!

Final Practice

Test Your IPv6 Subnetting Skills

Challenge 1 — Hospital Network

A hospital receives the prefix 2001:DB8:FACE::/48.

They need subnets for: Emergency, ICU, Radiology, Admin, Patient WiFi.

Q1: Network address for the Emergency subnet (Subnet 1)?

Q2: Network address for Patient WiFi (Subnet 5)?

Q3: Router address on the Admin subnet (Subnet 4)?

Challenge 2 — ISP Allocation

An ISP assigns the prefix 2001:DB8:ABCD:EF00::/56 to a business customer.

Q1: How many /64 subnets can the customer create?

Q2: What is the network address of their 20th subnet? (Hint: 20 decimal = 14 hex)

Q3: What is the first valid host address on the 20th subnet?

Final Quiz

Q1: In IPv6 subnetting, the Interface ID is always how many bits?

32 bits 48 bits 64 bits It varies

Q2: An organisation with a /48 prefix can create how many /64 subnets?

256 4,096 65,536 16,777,216

Final Quiz (continued)

Q3: Which part of the IPv6 address do you change when creating subnets from a /48?

The 1st hextet The 3rd hextet The 4th hextet (Subnet ID) The last hextet (Interface ID)

Q4: What is the main difference between IPv4 and IPv6 subnetting?

IPv6 requires more complex binary calculations IPv6 has a fixed 64-bit host part, making subnetting simpler IPv6 subnetting uses decimal numbers

Week 8 — Summary

Key Concepts

IPv6 addresses have 3 parts: Prefix, Subnet ID, Interface ID
The Interface ID is always 64 bits — never borrow from it
Subnets are always /64
Formula: subnets = 2^{(64 − prefix)}
Subnetting = change the Subnet ID (4th hextet for /48)

Common Prefix Sizes

/32 — Large ISP (4B subnets)
/48 — Enterprise / organisation (65K subnets)
/56 — Home / small business (256 subnets)
/64 — Single subnet (no further subnetting)

Next Week

Transport Layer — TCP vs UDP

Generating PDF…

Slides

Introduction to Networks

Week 8 — IPv6 Subnetting

Quick Review

Week 6 — IPv6 Fundamentals Recap

IPv6 Address — Quick Recap

IPv6 Address Types

Review Quiz

Review Quiz (continued)

IPv6 Simplification Practice

IPv6 Simplification Rules

IPv6 Subnetting

Dividing an IPv6 Network into Smaller Subnets

IPv4 vs IPv6 Subnetting

IPv4 Subnetting

IPv6 Subnetting

IPv6 Address Structure

The /48 Allocation

How IPv6 Subnetting Works

Hexadecimal — Quick Refresher

Hex ↔ Binary Practice

Example 1

Subnetting a /48 into /64 Subnets

Example 1 — The Scenario

Step 1 — Identify the Parts

Step 2 — Assign Subnet IDs

Step 3 — Assign Router Addresses

Your Turn — Example 1 Practice

Example 2

Subnetting with a /32 Prefix

Example 2 — Larger Allocation

Example 2 — Creating Subnets

The Simple Formula

Practice — How Many Subnets?

Example 3

School Network — Step by Step Subnetting

Example 3 — School Network

Example 3 — Solution

Your Turn — NZSE Network

Nibble Boundaries

Example 4

Home Network with a /56 Prefix

Example 4 — Home Network

Your Turn — /56 Practice

Common Mistakes to Avoid

❌ Don’t Do This

✅ Do This Instead

IPv4 vs IPv6 Subnetting — Side by Side

Final Practice

Test Your IPv6 Subnetting Skills

Challenge 1 — Hospital Network

Challenge 2 — ISP Allocation

Final Quiz

Final Quiz (continued)

Week 8 — Summary

Key Concepts

Common Prefix Sizes

Next Week