Virtualization Technology News and Information
Network Address Translation is a historically inefficient process for carriers. Until now.

By Hannes Gredler, CTO at RtBrick

Are you familiar with IPv4? It's the language used to navigate the internet and go from website to website. When it launched in 1983, only 4.3 billion unique addresses were allocated. They were quickly exhausted, with the regional Internet registry for Europe, RIPE NCC, allocating the last IPv4 addresses in 2019. No one anticipated how exponentially fast the number of internet-connected users and devices would grow. In 2024, the average US household has 17 internet-connected devices, more than imaginable 40 years ago.

The introduction of IPv6, the next iteration in the IP addressing series, solved this problem. However, a lot of equipment still only supports IPv4, and Cloudflare reported that only a third of IPv6-capable requests globally were made over IPv6 last year. So, although IPv6 helped with IPv4 exhaustion, it's still not being used as widely. 

IPv4 continues to own the largest share of IP address titles and will do so for the foreseeable future. As a result, operators widely depend on Carrier-Grade Network Address Translation (CGNAT), or the ability to translate private IPv4 addresses into public ones at a large scale. Think of network address translation a little like an apartment building in a packed city, which can house thousands of people at the same address rather than giving each person their own street and house number. The same concept can be applied to CGNAT.

Unfortunately, traditional approaches to this process come with several downsides, and it's prevented carriers from true cost, power, and space efficiency. Until now.

Traditional network address translation for carriers   

There are a couple of different ways that carriers can conduct address translation. First is a traditional two-box approach with a Broadband Network Gateway (BNG) and separate CGNAT device. This consumes more power and space, is more operationally complex, and there's a greater chance of a box failing. Carriers must navigate spares and maintenance processes for each box, and neither can be repurposed, impacting sustainability. E-waste reached a new high of about 136.6B pounds (62B kg) in 2022, and traditional two-box CGNAT has only exacerbated the issue. Now is the time to look for alternative options, especially with 2030 climate commitments just six years out.

The second approach is performing NAT on a card inside a large chassis-based BNG router. This method entails locked-in hardware, and there's an additional specialist line-card required - both of which still means greater costs and required power. It also doesn't enable carriers to scale down to smaller sites, making in-demand tasks like edge-computing more difficult.

Disaggregation: the solution carriers always needed

Disaggregation has opened a world of opportunities within our networks. Using powerful, cost-effective bare-metal switches, the approach enables software and hardware flexibility rather than locking carriers into one provider, which is the case of closed monolithic routing systems. The result is much leaner and more efficient systems.

Excitingly, as of February 2024, network address translation can also be disaggregated. CGNAT can now be enabled on Broadcom's Q2C chipset, so, for the first time, operators can use disaggregated routing software to deploy in-line CGNAT and a BNG on a single open switch. This was a significant advancement for carriers, who now get to enjoy reduced costs, power and space efficiency, simplified operations, and greener processes.

Building better networks for the future

CGNAT is just one example of how disaggregation is building better networks, and its deployment has picked up speed recently. By taking an open approach, operators are unlocking more power-efficient, reliable, and cost-effective broadband access that's previously been unattainable. They're also finding it easier to meet their 2024 company goals - from sustainability initiatives to enhanced security. Keep your eyes out for the next groundbreaking step disaggregation takes in could be around the corner!



hannes gredler 

As company founder and CTO, Hannes leads the vision and direction of RtBrick. He has 20+ years of expertise in engineering and supporting roles working with Alcatel (now Nokia Networks) and Juniper Networks. Hannes is also a co-author and contributor to multiple Internet Engineering Task Force (IETF) drafts and is a regular speaker at industry events and conferences. He holds 20+ patents in the IP multi-protocol label switching (IP/MPLS) space. 
Published Monday, April 22, 2024 7:33 AM by David Marshall
Filed under: ,
There are no comments for this post.
To post a comment, you must be a registered user. Registration is free and easy! Sign up now!
<April 2024>