Tickets - In Boston - higher latency to usbos01 than to usewr01 :: SixXS

Ticket ID: SIXXS #473654
Ticket Status: Invalid

PoP: usbos01 - OCCAID Inc. (Boston, Massachusetts)

In Boston - higher latency to usbos01 than to usewr01

Shadow Hawkins on Monday, 22 January 2007 16:53:36

From two different ADSL connections in Boston, I get lower latency to usewr01 than to usbos01. From one:

PING usbos01.sixxs.net (216.93.250.26) 56(84) bytes of data. 64 bytes from usbos01.sixxs.net (216.93.250.26): icmp_seq=1 ttl=53 time=22.5 ms 64 bytes from usbos01.sixxs.net (216.93.250.26): icmp_seq=2 ttl=53 time=22.8 ms 64 bytes from usbos01.sixxs.net (216.93.250.26): icmp_seq=3 ttl=53 time=22.3 ms 64 bytes from usbos01.sixxs.net (216.93.250.26): icmp_seq=4 ttl=53 time=22.4 ms --- usbos01.sixxs.net ping statistics --- 4 packets transmitted, 4 received, 0% packet loss, time 3002ms rtt min/avg/max/mdev = 22.380/22.536/22.837/0.258 ms PING usewr01.sixxs.net (66.55.128.25) 56(84) bytes of data. 64 bytes from usewr01.sixxs.net (66.55.128.25): icmp_seq=1 ttl=55 time=18.0 ms 64 bytes from usewr01.sixxs.net (66.55.128.25): icmp_seq=2 ttl=55 time=17.9 ms 64 bytes from usewr01.sixxs.net (66.55.128.25): icmp_seq=3 ttl=55 time=18.2 ms 64 bytes from usewr01.sixxs.net (66.55.128.25): icmp_seq=4 ttl=55 time=18.1 ms --- usewr01.sixxs.net ping statistics --- 4 packets transmitted, 4 received, 0% packet loss, time 2997ms rtt min/avg/max/mdev = 17.910/18.086/18.201/0.177 ms

From the other:

PING usbos01.sixxs.net (216.93.250.26) 56(84) bytes of data. 64 bytes from usbos01.sixxs.net (216.93.250.26): icmp_seq=1 ttl=49 time=17.0 ms 64 bytes from usbos01.sixxs.net (216.93.250.26): icmp_seq=2 ttl=49 time=17.7 ms 64 bytes from usbos01.sixxs.net (216.93.250.26): icmp_seq=3 ttl=49 time=17.3 ms 64 bytes from usbos01.sixxs.net (216.93.250.26): icmp_seq=4 ttl=49 time=18.4 ms --- usbos01.sixxs.net ping statistics --- 4 packets transmitted, 4 received, 0% packet loss, time 3002ms rtt min/avg/max/mdev = 17.044/17.644/18.426/0.540 ms PING usewr01.sixxs.net (66.55.128.25) 56(84) bytes of data. 64 bytes from usewr01.sixxs.net (66.55.128.25): icmp_seq=1 ttl=52 time=12.4 ms 64 bytes from usewr01.sixxs.net (66.55.128.25): icmp_seq=2 ttl=52 time=12.0 ms 64 bytes from usewr01.sixxs.net (66.55.128.25): icmp_seq=3 ttl=52 time=12.7 ms 64 bytes from usewr01.sixxs.net (66.55.128.25): icmp_seq=4 ttl=52 time=11.8 ms --- usewr01.sixxs.net ping statistics --- 4 packets transmitted, 4 received, 0% packet loss, time 3003ms rtt min/avg/max/mdev = 11.848/12.285/12.734/0.370 ms

But from MIT's connection, which is presumably better-peered than my home ADSL and my school's ADSL:

PING usbos01.sixxs.net (216.93.250.26) 56(84) bytes of data. 64 bytes from usbos01.sixxs.net (216.93.250.26): icmp_seq=1 ttl=54 time=4.21 ms 64 bytes from usbos01.sixxs.net (216.93.250.26): icmp_seq=2 ttl=54 time=8.51 ms 64 bytes from usbos01.sixxs.net (216.93.250.26): icmp_seq=3 ttl=54 time=6.88 ms 64 bytes from usbos01.sixxs.net (216.93.250.26): icmp_seq=4 ttl=54 time=6.10 ms --- usbos01.sixxs.net ping statistics --- 4 packets transmitted, 4 received, 0% packet loss, time 3057ms rtt min/avg/max/mdev = 4.212/6.428/8.514/1.550 ms PING usewr01.sixxs.net (66.55.128.25) 56(84) bytes of data. 64 bytes from usewr01.sixxs.net (66.55.128.25): icmp_seq=1 ttl=53 time=17.5 ms 64 bytes from usewr01.sixxs.net (66.55.128.25): icmp_seq=2 ttl=53 time=19.0 ms 64 bytes from usewr01.sixxs.net (66.55.128.25): icmp_seq=3 ttl=53 time=15.1 ms 64 bytes from usewr01.sixxs.net (66.55.128.25): icmp_seq=4 ttl=53 time=14.0 ms --- usewr01.sixxs.net ping statistics --- 4 packets transmitted, 4 received, 0% packet loss, time 3057ms rtt min/avg/max/mdev = 14.086/16.456/19.026/1.959 ms

Traceroute suggests that the problem stems from the fact that both of the ADSL connections are routed from Boston to New York and back before reaching the Boston PoP, while MIT's connection never leaves Boston. Is there a way to fix this (i.e., lower the latency), or am I just stuck with the circuitous routing that my ISP has set up?

State change: invalid Locked

Jeroen Massar SixXS Staff

on Monday, 22 January 2007 17:49:40

The state of this ticket has been changed to invalid

In Boston - higher latency to usbos01 than to usewr01

Jeroen Massar SixXS Staff

on Monday, 22 January 2007 17:51:44

We are not going to bother with a few milliseconds of latency. Latencies are dependent on network topology, the state of the moon etc, we can't do anything about that. Please don't open useless tickets.

In Boston - higher latency to usbos01 than to usewr01

Shadow Hawkins on Monday, 22 January 2007 21:20:11

Sorry, it just seemed strange to me.

Please note Posting is only allowed when you are logged in.