Re:Ask HN: Is not running a “dual stack” IPv4/IPv6 website c...
by 0xNOTVALID on Tuesday August 08, 2023 @01:41AM
(#37044933)
Can you support your statement that IPv6 has measurably better performance for websites?
Re:Ask HN: Is not running a “dual stack” IPv4/IPv6 website c...
by p1mrx on Tuesday August 08, 2023 @02:11AM
(#37045138)
This was claimed by Facebook in 2015: https://www.internetsociety.org/blog/2015/04/facebook-news-f...
Re:Ask HN: Is not running a “dual stack” IPv4/IPv6 website c...
by Dagger2 on Tuesday August 08, 2023 @09:27AM
(#37048663)
Apple also made a claim that v6 connection setup is 40% faster: https://www.zdnet.com/article/apple-tells-app-devs-to-use-ip...
(Not 1.4x faster overall, as you might think from the headline...)
These are two different things.
If your site is slow, it will cost you users. v6 has measurably better performance for websites, so not having it will cost you users. Those users will still be able to access your site though, they'll just choose not to because it's too slow for them.
The practical reality is that almost all client devices that are used for browsing the web will have some mechanism to reach legacy v4-only websites (but the mechanism might itself be a cause of poor performance...).
On the server side, v6-only servers (VPSs/containers/etc) are getting somewhat more common though, so if you're hosting something that's expected to be downloaded directly onto a server then please figure out v6 somehow.