The Internet Protocol (IP) is what gives every device on a network an address and allows packets to be routed between networks. IPv4 has been the workhorse of the internet since the 1980s, but its address space is exhausted — there simply aren’t enough 32-bit addresses for the number of connected devices in the world today. IPv6 was designed to solve this and several other structural limitations. This article walks through both protocols from first principles and explains, in depth, exactly how they differ.
What Is IPv4?
IPv4 uses a 32-bit address, written in dotted-decimal notation, e.g., 192.168.1.10. This gives a theoretical maximum of about 4.3 billion unique addresses — which sounded like plenty in the 1980s but has long since been exhausted at the global allocation level.
IPv4 Address Structure
An IPv4 address is divided into a network portion and a host portion, determined by a subnet mask:
IP Address: 192.168.1.10
Subnet Mask: 255.255.255.0 (/24)
Network: 192.168.1.0
Host Range: 192.168.1.1 - 192.168.1.254
Broadcast: 192.168.1.255
What Is IPv6?
IPv6 uses a 128-bit address, written in hexadecimal and separated by colons, e.g.:
2001:0db8:85a3:0000:0000:8a2e:0370:7334
This can be abbreviated by removing leading zeros in each group and replacing one run of consecutive all-zero groups with :::
2001:db8:85a3::8a2e:370:7334
128 bits provides approximately 340 undecillion addresses (3.4 × 10^38) — enough to assign a unique address to every grain of sand on Earth many times over, and then some.
Side-by-Side Comparison
| Feature | IPv4 | IPv6 |
|---|---|---|
| Address length | 32 bits | 128 bits |
| Address format | Dotted decimal (192.168.1.1) | Hexadecimal, colon-separated |
| Total addresses | ~4.3 billion | ~340 undecillion |
| Header size | 20–60 bytes (variable, with options) | Fixed 40 bytes |
| Header complexity | Includes checksum, options | Simplified; extension headers instead |
| Broadcast | Yes | No — replaced by multicast |
| Address configuration | Manual, DHCP | Manual, DHCPv6, SLAAC (auto-config) |
| Fragmentation | Done by routers and sender | Only by sender (path MTU discovery) |
| Built-in security (IPsec) | Optional | Originally mandatory in spec, now optional in practice |
| NAT requirement | Common, due to address scarcity | Generally unnecessary due to abundant address space |
| Header checksum | Present | Removed (handled at Layer 2/4 instead) |
| Address Resolution | ARP | Neighbor Discovery Protocol (NDP) using ICMPv6 |
Key Structural Differences Explained
1. Address Space
This is the headline difference. IPv4’s exhaustion is why IPv6 exists. Techniques like NAT and CIDR extended IPv4’s life by decades, but they are workarounds, not solutions — NAT in particular breaks the original end-to-end connectivity model of the internet.
2. Header Design
The IPv4 header is variable-length and includes a header checksum that every router must recalculate at every hop, adding processing overhead. The IPv6 header is a fixed 40 bytes with no checksum — error checking is left to Layer 2 (Ethernet FCS) and Layer 4 (TCP/UDP checksums), making IPv6 packet processing faster and simpler for routers.
graph TB
subgraph IPv4 Header
A1[Version/IHL] --> A2[Type of Service]
A2 --> A3[Total Length]
A3 --> A4[Fragmentation Fields]
A4 --> A5[TTL/Protocol/Checksum]
A5 --> A6[Source/Dest Address 32-bit]
end
subgraph IPv6 Header
B1[Version/Traffic Class/Flow Label] --> B2[Payload Length]
B2 --> B3[Next Header/Hop Limit]
B3 --> B4[Source/Dest Address 128-bit]
end3. Address Autoconfiguration
IPv4 devices typically need DHCP or manual configuration. IPv6 supports SLAAC (Stateless Address Autoconfiguration), where a device can generate its own address using the network prefix advertised by a router plus its own interface identifier — no DHCP server required, though DHCPv6 is still available for stateful configuration when needed.
4. No More Broadcast
IPv4 relies on broadcast addresses (e.g., 192.168.1.255) for things like ARP requests, which every device on the segment must process. IPv6 eliminates broadcast entirely in favor of multicast, so only interested devices process the traffic — this is more efficient, especially on large networks.
5. Neighbor Discovery Protocol (NDP) Replaces ARP
IPv4 uses ARP (Address Resolution Protocol) broadcasts to map IP addresses to MAC addresses. IPv6 uses NDP, built on ICMPv6 multicast messages, to perform the same function along with router discovery and duplicate address detection.
6. Fragmentation
In IPv4, both the sending host and intermediate routers can fragment a packet that’s too large for the next link’s MTU. In IPv6, only the source host fragments — routers simply drop oversized packets and send back an ICMPv6 “Packet Too Big” message, pushing the sender to rely on Path MTU Discovery. This simplifies router processing significantly.
Practical Examples
Checking IP Configuration on Linux
# View IPv4 and IPv6 addresses
ip addr show
# Example output snippet
# inet 192.168.1.15/24 brd 192.168.1.255 scope global eth0
# inet6 2001:db8:0:1::15/64 scope global
# inet6 fe80::a00:27ff:fe4e:66a1/64 scope link
Notice the fe80:: address — this is a link-local address, automatically assigned to every IPv6 interface and used only for communication on the local segment (similar in spirit to APIPA’s 169.254.x.x in IPv4, but present by design rather than as a fallback).
Enabling a Static IPv6 Address on Linux (Netplan)
network:
version: 2
ethernets:
eth0:
addresses:
- 2001:db8:0:1::10/64
gateway6: 2001:db8:0:1::1
nameservers:
addresses: [2001:4860:4860::8888]
Configuring IPv6 on a Cisco Router
Router(config)# ipv6 unicast-routing
Router(config)# interface gigabitEthernet 0/0
Router(config-if)# ipv6 address 2001:db8:0:1::1/64
Router(config-if)# ipv6 enable
Router(config-if)# no shutdown
Router# show ipv6 interface brief
Dual-Stack Ping Test
ping -4 8.8.8.8
ping -6 2001:4860:4860::8888
Python: Checking Whether an Address Is IPv4 or IPv6
import ipaddress
def classify(addr):
ip = ipaddress.ip_address(addr)
return "IPv6" if ip.version == 6 else "IPv4"
test_addresses = ["192.168.1.1", "2001:db8::1", "10.0.0.5"]
for a in test_addresses:
print(f"{a}: {classify(a)}")
Transition Mechanisms
Because the internet cannot switch overnight, several coexistence mechanisms exist:
| Mechanism | Description |
|---|---|
| Dual Stack | Devices run both IPv4 and IPv6 simultaneously — the most common real-world approach |
| Tunneling (6to4, Teredo) | Encapsulates IPv6 packets inside IPv4 to cross IPv4-only networks |
| NAT64/DNS64 | Allows IPv6-only clients to reach IPv4-only services |
Best Practices
- Always deploy dual-stack when introducing IPv6 rather than a “flag day” cutover.
- Don’t assume NAT-equivalent security in IPv6 — since most devices get globally routable addresses, a properly configured firewall is essential, not optional.
- Use
/64as the standard subnet size for IPv6 LANs; it’s the recommended default for SLAAC to function correctly. - Filter ICMPv6 carefully — unlike IPv4, IPv6 depends on ICMPv6 for core functions like NDP and Path MTU Discovery, so don’t block it entirely.
- Update monitoring and logging systems to correctly parse and store IPv6 addresses.
Troubleshooting
| Symptom | Likely Cause | Fix |
|---|---|---|
| No IPv6 connectivity | Router not advertising prefix | Check show ipv6 interface brief / router advertisements |
| Only link-local address present | No RA (Router Advertisement) received | Verify ipv6 unicast-routing is enabled and RA is not suppressed |
| Dual-stack app prefers wrong protocol | Address selection policy | Check getaddrinfo order / /etc/gai.conf on Linux |
| IPv6 works locally but not to internet | ISP doesn’t support IPv6, or tunnel misconfigured | Confirm ISP IPv6 support, check tunnel broker config |
Further Reading
- RFC 8200 — Internet Protocol, Version 6 (IPv6) Specification
- RFC 791 — Internet Protocol (IPv4)
- IANA IPv4 Address Report
- Cisco IPv6 Configuration Guide
- Google IPv6 Adoption Statistics
Conclusion
IPv6 isn’t just “IPv4 with more addresses” — it rethinks the header format, address resolution, autoconfiguration, and fragmentation model to make routing more efficient and management simpler at internet scale. As IPv4 exhaustion continues to push adoption forward, understanding both protocols — and how to run them side by side — is a core skill for anyone working in networking today.
