IETF reviews draft to modernize IPv4 by replacing 40-year-old ARP
The IETF IntArea working group is evaluating a new draft that proposes replacing IPv4 ARP with IPv6 neighbor discovery to deliver IPv4 services over IPv6-only networks. By eliminating the reliance on subnets and FHRP, this architectural shift aims to simplify infrastructure management for streaming service providers and data centers.
Key Takeaways
- The 'v4-with-v6-nh' draft proposes using the IPv6 neighbor cache to resolve IPv4 addresses, removing the 40-year-old reliance on ARP.
- The architectural shift eliminates traditional /24 subnets in favor of flat /32 address pools, reducing address waste on broadcast and gateway overhead.
- Managed reference implementations have been developed for four operating system families, including Windows 11 and macOS.
- Operational testing shows the model enables active-active routing without the need for virtual MAC addresses or gateway election protocols like VRRP.
Why It Matters
For streaming providers and CDN operators, this draft offers a path to simplify data center fabrics by removing the parallel 'layer cake' of IPv4 scaffolding. By treating IPv4 as a service on top of an IPv6-only control plane, operators can eliminate subnet sizing constraints and renumbering risks. Concretely, this allows for more efficient IP asset utilization and faster network convergence through pure IGP machinery rather than brittle election protocols. Watch the official IETF IntArea mailing list for the working group's adoption decision, which would signal a major formalization of current 'off-link' practices used by large hosting providers.
Additional Context
The push to decouple IPv4 reachability from legacy local-network architecture comes as IPv6 adoption reached a historic tipping point in 2026. In March 2026, native IPv6 traffic on Google’s network exceeded 50% for the first time, signaling that for more than half of global internet users, legacy IPv4 is no longer the primary transport protocol. However, per data from early 2026, roughly 76% of Fortune 500 companies still operate primarily on IPv4 infrastructure, creating a massive long-tail demand for IPv4 services even as core networks move toward IPv6-only designs.
This gap between protocol usage and infrastructure capability has turned IPv4 addresses into strategic assets. According to an April 2026 market report, /24 blocks are currently trading between $35 and $45 per IP, leading operators to seek technical solutions that reduce address waste. Legacy subnetting architectures typically lose several addresses per block to network and broadcast requirements; moving to the /32 model proposed in the IETF draft allows providers to recover those addresses. Larger hosting entities like Hetzner and OVHcloud have used similar undocumented workarounds for years to maximize their IP stock.
Simultaneously, client-side ecosystems are hardening their IPv6-first postures. Microsoft recently enabled CLAT (Customer-side Translator) by default in Windows 11, following a broader industry trend toward 'IPv6-mostly' environments. Per a February 2026 assessment by Cisco, approximately 62% of traffic in modern conference environments now moves via IPv6, leaving IPv4 to function essentially as a compatibility layer for legacy applications. The IETF IntArea draft represents the final step in this evolution: removing the 1982-era ARP mechanism to let IPv4 run natively on modern IPv6 stacks without translation or tunneling.
Read full article at labs.ripe.net
Enjoy our coverage?
Add StreamingMeme as a preferred source on Google to see more of our streaming news at the top of your Search results.
Add as preferred source