I'm looking for a general explanation of IPv6 addressing architecture and subnetting mechanisms. Specifically, what are the advantages of having a 128-bit address length, how does CIDR-based subnetting and route summarization work? How does IPv6 manage limited address spaces in network planning, and what are common practical approaches? What resources and sources do you recommend for this topic, and what are your experiences with it?
How does IPv6 addressing and subnetting work?
👁️ 11 views💬 1 replies❤️ 0 likes
1 Replies
IPv6 expands the address length from 32 bits to 128 bits, fundamentally solving the IPv4 address exhaustion problem while also providing more room for route aggregation. Each 128-bit address is divided into a network prefix and an interface identifier (IID), with a common allocation method being a 64-bit prefix + 64-bit IID. The direct benefit of this approach is that it allows endpoints to generate their own IID using SLAAC (Stateless Address Autoconfiguration) without compromising global uniqueness, reducing the configuration burden of DHCPv6.
In subnet division, IPv6 fully adheres to the CIDR concept: adjusting the prefix length to determine subnet scale. Enterprises typically use /48 or /56 as site prefixes, further divided into /64 subnets, since /64 is the default subnet size assumed by most protocols (e.g., ND, RA). Route aggregation between routers is highly efficient—if prefixes are contiguous, they can be replaced with a single shorter prefix (e.g., aggregating multiple /64s into a /48), significantly reducing the number of entries in the global routing table. Thanks to the low cost of route aggregation, ISPs often only advertise prefixes at a few hierarchical levels in their backbone networks, keeping the routing scale of the entire internet at around tens of thousands of entries.
In practical planning, a common approach is to first define prefixes for sites, departments, and data centers in an address planning tool (e.g., IPv6 Address Planner, Cisco Prime), then uniformly allocate /64 subnets to each VLAN or VPC. For scenarios requiring finer-grained isolation (e.g., multi-tenant cloud platforms), subnet division can be extended to /72, /80, or even /96 above /64, combined with firewall policies for precise control. Recommended reading includes RFC 4291 (IPv6 Addressing Architecture), RFC 6177 (IPv6 Address Allocation Policy), and IETF’s "IPv6 Addressing Architecture" document; additionally, Cisco and Juniper’s official whitepapers and *IPv6 Fundamentals* are excellent resources for beginners and practitioners. Hopefully, these insights help you quickly build a scalable IPv6 network.