I saw this in a post from Travis Wise of Google yesterday. Pretty cool for
those users who do not want to use their ISP's name servers, or just want to
have dns resolve quickly from anywhere in the world. In either case, I think
it is cool ;-]
http://code.google.com/speed/public-dns/
Here is the original post - Yes, this one is public... oops!
https://plus.google.com/111937447827665620879/posts/27S6QB8j1Ry
Nice easy numbers to remember too. 8.8.8.8 and 8.8.4.4
This service has been discussed several times in the ~2 years since it was
first released (including topics such as why it's bad for CDNs)
The archives would be a good place to start...
Scott.
Todd Underwood wrote:
not bad for CDNs anymore:
http://arstechnica.com/telecom/news/2011/08/opendns-and-google-working-with-cdns-on-dns-speedup.ars
t
Fwiw, ol' Steve Gibson has written a small (167KB), .exe, "DNS Benchmark".
It's easy to add 8.8.8.8 and 8.8.8.4 (or any nameserver) to the .ini file from within the program .
http://www.grc.com/dns/benchmark.htm
--Michael
Wow, consider me educated on old news. LOL
I imagine it is new to many of the users on the new service they launched. I
completely forgot that Nanog would probably be the first to comment and
chime in when it first became available.
Thanks for the information. I will definitely research the CDN issues. Back
to school steve...Why did I
Awesome link Todd - Why did I think that the resolving server would already
know "where network path wise" the request came from. Let me post this as a
comment and ask how the CDN endpoint routing is working.
Better yet, change the dns on my router, and Netflix on the Google TV should
let me know how well it is streaming.
--steve
I see those guys at BBNPlanet is trying to compete with them using 4.2.2.1, .2, and .3.
johno
C'mon, John; no trolling. 
Cheers,
-- jra
I would guess, using this -
http://tools.ietf.org/html/draft-vandergaast-edns-client-subnet-00
Note the authors (two from Google), and the initial release date (not
actually shown in the that version as far as I can see, but it was around
the same time Google announced their public DNS servers).
Scott.
jan 27 2011, so says the doc header...
There's also namebench, does a lot of more tests, and runs at Mac OSX and Linux too: http://code.google.com/p/namebench/
Michiel Klaver wrote:
Fwiw, ol' Steve Gibson has written a small (167KB), .exe, "DNS Benchmark".
It's easy to add 8.8.8.8 and 8.8.8.4 (or any nameserver) to the .ini file
from within the program .
http://www.grc.com/dns/benchmark.htm
--Michael
There's also namebench, does a lot of more tests, and runs at Mac OSX and
Linux too: http://code.google.com/p/namebench/
More tests? Where's the result of the DNSSec checks?
Its maintenance is suspect, since my ISP's (and most resolvers) returned something like:
a.. www.anonymizer.com appears incorrect: 209.143.153.58
a.. isohunt.com appears incorrect: 208.95.172.130
a.. www.thesouthasian.org appears incorrect: sbsfe.geo.mf0.yahoodns.net
a.. youporn.com appears incorrect: 173.192.24.120, 173.192.60.242, 173.192.60.245, 173.192.24.114, 173.192.24.115, 173.192.24.116, 173.192.24.117, 173.192.24.119
a.. www.stopkinderporno.com appears incorrect: 188.72.230.78
a.. wikileaks.org appears incorrect: 88.80.16.63
a.. www.lapsiporno.info appears incorrect: 89.166.50.123
a.. www.paypal.com is hijacked: 173.0.88.34, 173.0.84.2, 173.0.84.34, 173.0.88.2
a.. uddthailand.com appears incorrect: 184.173.208.195
a.. www.stormfront.org appears incorrect: 174.121.229.156
a.. motherless.com appears incorrect: 198.64.4.17, 198.64.4.16
a.. www.partypoker.com appears incorrect: ppdotcom.iivt.com
a.. twitter.com appears incorrect: 199.59.149.198, 199.59.149.230, 199.59.148.10
Interesting choice of URLs.
I wonder how many folks are wasting their time chasing this ominous sounding
a.. www.paypal.com is hijacked: 173.0.88.34, 173.0.84.2, 173.0.84.34, 173.0.88.2
--Michael
I guess you selected the Alexa top1000 as data-source, which contains this selection of URLs.
The result of mis-matching IP addresses reports could be the result of geo-dns, serving different results to different parts of the world to match local CDN nodes.
The original draft had a different name, and was released in Jan 2010.
http://tools.ietf.org/html/draft-vandergaast-edns-client-ip-00
Scott