←back to thread

1895 points _l4jh | 1 comments | | HN request time: 0s | source
Show context
cleanbrowsing ◴[] No.16729310[source]
And look at these ping times:

                                   CloudFlare       Google DNS       Quad9            OpenDNS          
  NewYork                            2 msec           1 msec           2 msec           19 msec          
  Toronto                            2 msec           28 msec          17 msec          27 msec          
  Atlanta                            1 msec           2 msec           1 msec           19 msec          
  Dallas                             1 msec           9 msec           1 msec           7 msec           
  San Francisco                      3 msec           21 msec          15 msec          20 msec          
  London                             1 msec           12 msec          1 msec           14 msec          
  Amsterdam                          2 msec           6 msec           1 msec           6 msec           
  Frankfurt                          1 msec           9 msec           2 msec           9 msec           
  Tokyo                              2 msec           2 msec           81 msec          77 msec          
  Singapore                          2 msec           2 msec           1 msec           189 msec         
  Sydney                             1 msec           130 msec         1 msec           165 msec

Very impressive CloudFlare.
replies(20): >>16729423 #>>16729467 #>>16729545 #>>16729560 #>>16729939 #>>16729952 #>>16730034 #>>16730110 #>>16730198 #>>16730229 #>>16730567 #>>16730893 #>>16731389 #>>16732068 #>>16732273 #>>16732936 #>>16733149 #>>16733462 #>>16733833 #>>16761330 #
chrissnell ◴[] No.16729467[source]
Where are you testing from? I'm going to guess: a datacenter. Residential customers won't see anything this fast. I'm in a small town in Kansas, connected by 1 Gbit ATT fiber. I'm getting ~26ms to 1.1.1.1 and ~19ms to my private DNS resolver that I host in a datacenter in Dallas. Google DNS comes in around 19ms.

I suspect that Cloudflare and Google DNS both have POPs in Dallas, which accounts for the similar numbers to my private resolver. My point is, low latencies to datacenter-located resolver clients is great but the advantage is reduced when consumer internet users have to go across their ISP's long private fiber hauls to get to a POP. Once you're at the exchange point, it doesn't really matter which provider you choose. Go with the one with the least censorship, best security, and most privacy. For me, that's the one I run myself.

Side note: I wish AT&T was better about peering outside of their major transit POPs and better about building smaller POPs in regional hubs. For me, that would be Kansas City. Tons of big ISPs and content providers peer in KC but AT&T skips them all and appears to backhaul all Kansas traffic to DFW before doing any peering.

replies(9): >>16729476 #>>16730076 #>>16730117 #>>16730356 #>>16731306 #>>16731480 #>>16732326 #>>16732414 #>>16733837 #
matthberg ◴[] No.16731480[source]
Ping from University of Rochester, over wifi:

Cloudflare:

  64 bytes from 1.1.1.1: icmp_seq=0 ttl=128 time=2 ms
  64 bytes from 1.1.1.1: icmp_seq=1 ttl=128 time=2 ms
  64 bytes from 1.1.1.1: icmp_seq=2 ttl=128 time=2 ms
  64 bytes from 1.1.1.1: icmp_seq=3 ttl=128 time=9 ms
  64 bytes from 1.1.1.1: icmp_seq=4 ttl=128 time=2 ms
Google:

  64 bytes from 8.8.8.8: icmp_seq=0 ttl=54 time=12 ms
  64 bytes from 8.8.8.8: icmp_seq=1 ttl=54 time=11 ms
  64 bytes from 8.8.8.8: icmp_seq=2 ttl=54 time=13 ms
  64 bytes from 8.8.8.8: icmp_seq=3 ttl=54 time=45 ms
  64 bytes from 8.8.8.8: icmp_seq=4 ttl=54 time=14 ms
  64 bytes from 8.8.8.8: icmp_seq=5 ttl=54 time=11 ms
  64 bytes from 8.8.8.8: icmp_seq=6 ttl=54 time=34 ms
Quad9:

  64 bytes from 9.9.9.9: icmp_seq=0 ttl=53 time=10 ms
  64 bytes from 9.9.9.9: icmp_seq=1 ttl=53 time=69 ms
  64 bytes from 9.9.9.9: icmp_seq=2 ttl=53 time=14 ms
  64 bytes from 9.9.9.9: icmp_seq=3 ttl=53 time=58 ms
  64 bytes from 9.9.9.9: icmp_seq=4 ttl=53 time=52 ms
One thing I noticed is that when I first pinged 1.1.1.1 I got 14ms, which then quickly dropped to ~3ms consistently:

  64 bytes from 1.1.1.1: icmp_seq=0 ttl=128 time=14 ms
  64 bytes from 1.1.1.1: icmp_seq=1 ttl=128 time=14 ms
  64 bytes from 1.1.1.1: icmp_seq=2 ttl=128 time=2 ms
  64 bytes from 1.1.1.1: icmp_seq=3 ttl=128 time=3 ms
  64 bytes from 1.1.1.1: icmp_seq=4 ttl=128 time=1 ms
  64 bytes from 1.1.1.1: icmp_seq=5 ttl=128 time=4 ms
replies(6): >>16731753 #>>16731779 #>>16731866 #>>16732407 #>>16732515 #>>16733059 #
Tree1993 ◴[] No.16731753[source]
Beijing:

  PING 1.1.1.1 (1.1.1.1): 56 data bytes
  64 bytes from 1.1.1.1: icmp_seq=0 ttl=52 time=241.529 ms
  64 bytes from 1.1.1.1: icmp_seq=1 ttl=52 time=318.034 ms
  64 bytes from 1.1.1.1: icmp_seq=2 ttl=52 time=337.291 ms
  64 bytes from 1.1.1.1: icmp_seq=3 ttl=52 time=255.748 ms
  64 bytes from 1.1.1.1: icmp_seq=4 ttl=52 time=247.765 ms
  64 bytes from 1.1.1.1: icmp_seq=5 ttl=52 time=235.611 ms
  64 bytes from 1.1.1.1: icmp_seq=6 ttl=52 time=239.427 ms
  64 bytes from 1.1.1.1: icmp_seq=7 ttl=52 time=247.911 ms
  64 bytes from 1.1.1.1: icmp_seq=8 ttl=52 time=260.911 ms
  64 bytes from 1.1.1.1: icmp_seq=9 ttl=52 time=281.153 ms
  64 bytes from 1.1.1.1: icmp_seq=10 ttl=52 time=300.363 ms
  64 bytes from 1.1.1.1: icmp_seq=11 ttl=52 time=234.296 ms
replies(4): >>16732175 #>>16732615 #>>16732754 #>>16732978 #
1. oldsharp ◴[] No.16732615[source]
Hangzhou:

    $ ping 1.1.1.1
    PING 1.1.1.1 (1.1.1.1): 56 data bytes
    Request timeout for icmp_seq 0
    Request timeout for icmp_seq 1
    Request timeout for icmp_seq 2
    Request timeout for icmp_seq 3
    Request timeout for icmp_seq 4
    Request timeout for icmp_seq 5
    Request timeout for icmp_seq 6
    Request timeout for icmp_seq 7
    Request timeout for icmp_seq 8
    Request timeout for icmp_seq 9
    Request timeout for icmp_seq 10

    $ ping 1.0.0.1
    PING 1.0.0.1 (1.0.0.1): 56 data bytes
    64 bytes from 1.0.0.1: icmp_seq=0 ttl=50 time=167.359 ms
    64 bytes from 1.0.0.1: icmp_seq=1 ttl=50 time=165.791 ms
    64 bytes from 1.0.0.1: icmp_seq=2 ttl=50 time=165.846 ms
    64 bytes from 1.0.0.1: icmp_seq=3 ttl=50 time=166.755 ms
    64 bytes from 1.0.0.1: icmp_seq=4 ttl=50 time=166.694 ms
    64 bytes from 1.0.0.1: icmp_seq=5 ttl=50 time=166.088 ms
    64 bytes from 1.0.0.1: icmp_seq=6 ttl=50 time=166.460 ms
    64 bytes from 1.0.0.1: icmp_seq=7 ttl=50 time=166.668 ms
    64 bytes from 1.0.0.1: icmp_seq=8 ttl=50 time=166.753 ms
    64 bytes from 1.0.0.1: icmp_seq=9 ttl=50 time=165.670 ms
    64 bytes from 1.0.0.1: icmp_seq=10 ttl=50 time=166.816 ms
Seem not China friendly :-(