Last month, APNIC arranged and held a panel session, IPv6 is Live! at the Pacific Telecommunications Council’s annual conference PTC’17.
Panellists included guest speakers from Telekom Malaysia, Telstra, Comcast, LinkedIn, and 6connect, all of whom have played leading roles in deploying IPv6 on their networks, and in turn, have made significant contributions to improving the IPv6 capabilities in their economies and industries.
Growing networks, v4 exhaustion: key reasons for move to v6
All panellists pointed to IANA’s final allocation of IPv4 addresses from the ‘free pool’ in 2011 as a major factor which kickstarted, or in the case of Comcast and 6connect, justified their IPv6 deployment plans.
“The main reason we chose to do IPv6 was IPv4 exhaustion, which affected not only our public but private addresses as well,” said Sunny Yeung, a Senior Technology Specialist at Telstra, who recently shared some lessons from Telstra’s IPv6 deployment, which was five years in the making.
Sunny went on to say the decision was justified further as a means of reducing future capital expenditure, a comment reiterated by each of the other panellists. “If you want to spend money, you want to spend money that results in growth, so why would you spend money on extending your IPv4 resources through NAT translations rather than investing in new technologies that can accommodate [future services like] IoT,” Sunny said.
Mobile leading the charge
As an introduction to the panel discussion, APNIC’s Chief Scientist, Geoff Huston, shared his analysis of the worldwide growth of IPv6 (the percentage of users accessing Google over IPv6, for example, is currently around 16%).
Geoff recognized the “huge amount of work” Comcast has done in this area in the USA — having provisioned 90% of its broadband customers with IPv6 — and highlighted recent growth generated by several very large carriers, including Reliance Jio in India and Sky in the UK.
Geoff predicts IPv6 growth in the mobile arena will continue to grow as providers increasingly recognize they cannot sustain the use of NAT technology.
“My suspicion is we have three to five years of NAT’ing and then it will break, forcing folk to go IPv6-only,” he said.
Adding to this theme from a content provider’s perspective, Zaid Ali Kahn, the Senior Director of Infrastructure Engineering at LinkedIn, noted they have recorded instances where IPv6 over mobile has performed better than IPv4.
“Because we own our mobile stack, we can measure whether our users are coming by IPv6 or IPv4,” Zaid said. “Using our measuring technique, we are seeing with certain providers better download times by IPv6 than IPv4. Why that is, we’re still trying to figure out. But measuring this gives us evidence to give to our mobile developers to continue to support v6.”
Dual stack is preferred deployment strategy, but single stack is end goal
All panellists said their strategy had involved deploying IPv6 using dual stack.
“Dual stack was a way that we could incrementally provide IPv6 support for broadband,” said John Brzozowski, Fellow and Chief Architect at Comcast Cable, which started working on deploying IPv6 over its network in 2005. He adds that although dual stack has provided their team and their customers with the greatest amount of flexibility in migrating to IPv6, the end goal is IPv6-only.
“On my tenth anniversary at Comcast in 2015, my boss asked what I want to do now that we’ve deployed v6,” recalled John. “I answered that I want to turn off v4, not just as an experiment but it’s the best thing we can do for the business, customers, and industry.”
“Around 40% of our Internet traffic is over IPv6. Once that passes 51% and continues to grow, IPv4 becomes the protocol that gets turned off.”
Zaid shared a similar story. “We wanted to go v6-only but we wanted to be realistic with our developers to make sure they were comfortable with the protocol. So, we said we’d dual stack our data centres over 2015-17. By 2018 we’ll turn off v4 in our data centres because we don’t want to be supporting two stacks.”
Sunny said he also had similar ideas for Telstra too. “Dual stack is an effective transition technology; however, it doesn’t solve your IPv4 depletion problem.”
“Our long-term goal is to go to single stack IPv6. If you don’t have a strategy to go to single stack IPv6 from day one, it’s going to be really hard when you transition from dual stack to single stack later. So, you need to look at your dual stack and single stack strategy as one.”
The views expressed by the authors of this blog are their own and do not necessarily reflect the views of APNIC. Please note a Code of Conduct applies to this blog.