I recently had the honor of being invited back to Paris by Eric Vyncke and Mark Townsley to present at the Cisco Ecole Polytechnique Networking Innovation and Research Symposium. I generally tend to avoid Paris in the winter but the opportunity was too good to pass up, especially with the amazing lineup of speakers.
Last year at a slightly different conference, I spent time talking about our IPv6 transition at LinkedIn. This year we are almost at the two-year mark running IPv6 on www.linkedin.com and I was eager to share some statistics on IPv6 growth and performance measurement.
One of the advantages of end-to-end control with a mobile product is the ability to do effective measurements at scale. I wanted to share our RUM-based measuring techniques and also encourage other folks in the content/application space to do so.
As a summary, we are seeing 10% adoption of IPv6 from our member base which is very much in line with what Google is reporting as well.
We spent a bit of time looking at performance and find that IPv6 is performing better than IPv4 on mobile networks. For some select networks in Europe, LinkedIn is seeing up to 40% performance improvements over IPv6, and in the US, up to 10%.
Other interesting data we are observing is the TCP timeout on IPv4 over mobile carrier networks is as high as 4.6% and IPv6 timeouts are on a much lower side at 1.6%. We are not sure what is causing this difference in performance but encourage mobile carriers to look into this and also welcome sharing data.
Overall, we are happy to report that there is increasing adoption of IPv6 and also better performance when visiting LinkedIn if you are visiting our site through mobile carrier networks.
The nice folks at Cisco Ecole Polytechnique Networking Innovation and Research Symposium have put my talk on YouTube so you can check it out and I also uploaded my presentation on SlideShare.
Original post appeared on LinkedIn Pulse
Zaid Ali Kahn is Senior Director, Global Infrastructure Architecture and Strategy, at LinkedIn.
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.