Can CGNAT handle the Internet of Things?

Preserving IPv4 Infrastructure and Transitioning to IPv6

The amount of devices that keep connecting to the Internet keeps on growing and there is no stopping it. Smartphones cannot even be called a trend anymore; the adoption keeps expanding rapidly. Even last year, half of the Americans already owned a smartphone and you know what that means—apart from socializing issues: it needs an IP address to connect to the Internet. And so does your TV, your online streaming device, your tablet, your laptop and soon it will be the fridge (if not already) and likely, the kitchen sink.

The problem is though, that we are effectively out of IP addresses, and organizations know that IPv6 promises to solve this problem. They are also very aware this at least raises some immediate questions (does my network infrastructure even support this?) and probably more issues that need looking into: do your end user devices support it, your applications? And then the biggest Driver of Things: money. What will it cost to move to IPv6?. “What if I just don’t?” is a valid question, at least from some peoples’ perspectives…

Carrier Grade NAT (CGNAT) is there to help keep organizations working with IPv4 and keep expanding connected devices. NAT has done a great job oversubscribing a public IP address with multiple private IPs in the internal network. This works great for web browsing but creates big issues for device-to-device communications: The IP protocol was designed to have devices connect to each other transparently. And NAT as found on firewalls, Wi-Fi routers and the likes, are very likely to break these communications.

CGNAT was designed not just to do NAT at large scale (older RFCs called the technology Large Scale NAT or “LSN”) but to do so in the most transparent way, with standardized behavior. But how far can we stretch out the little IPv4 addresses we have?

For quite a while, or so it seems. A10’s CGNAT solutions provide many ALGs (Application Layer Gateways) that provide a transparent communication for many applications, and we can do this at a very large scale in very compact form factors.

Fortunately, the Thunder Series Application Delivery Controllers (ADCs) (as well as the AX Series ADCs) have CGNAT there as a first step, with IPv6 transition technologies also included. CGNAT is set to alleviate the immediate IPv4 shortage quite easily. But our all-inclusive feature licensing and advanced IPv6 migration technologies make a phased approach to IPv6 easy and affordable.

Just how long CGNAT will stretch out the IPv4 usage remains to be seen. I would like to hear more opinions, please come and see me in the panel discussion (Carrier Grade NAT – Good, Bad or Ugly and Necessary?)  at Nov 13th, 1:15pm the gogoNET LIVE! 4 event where you can learn more about IPv6, and how to get there.

Paul Nicholson
November 7, 2013

About Paul Nicholson

Paul Nicholson brings 24 years of experience working with Internet and security companies in the U.S. and U.K. In his current position, Nicholson is responsible for global product marketing and strategy at San Jose, Calif.-based application networking and security leader A10 Networks. Prior to A10 Networks, Nicholson held various technical and management positions at Intel, Pandesic (the Internet company from Intel and SAP), Secure Computing, and various security start-ups. READ MORE