Regarding high latency from Los Angeles/surrounding area

[Rapture]

I do use Google DNS, but that only affects routing when you are dealing with a CDN or similar with multiple IPs servicing the same content.

In this case, the FFXIV world servers have static IPs, with routes determined by advertised BGP tables between core backbone routers and ISP endpoints. I checked the Looking Glass traceroutes, and it seems Telia, Global Crossing, and Sprint always go identical routes through as6453.net to the NA datacenter from my geographical location.

Level3 seems to go through Cogent, but more hops, which is likely why it's not preferred by BGP. This other route which goes through Cogent directly to 192.34.76.2 may be more rarely used for all I know. I'll see what the Looking Glass says regarding latency during peak time.

Worth noting that now that the servers are all down for maintenance and empty of users, the tracert is normal. Definitely suggests S-E may not have purchased enough capacity on the as6453.net backbone to handle the increased active users from their server upgrades these past few weeks:

1 <1 ms <1 ms <1 ms 192.168.1.254
2 1 ms 1 ms <1 ms 001.131-78-65.ftth.swbr.surewest.net [65.78.131.1]
3 1 ms 1 ms 1 ms 172.22.5.13
4 1 ms <1 ms <1 ms 172.21.2.13
5 1 ms <1 ms <1 ms 172.21.0.250
6 4 ms 4 ms 4 ms sjo-bb1-link.telia.net [213.248.88.73]
7 4 ms 4 ms 4 ms tata-ic-140123-sjo-bb1.c.telia.net [213.248.95.130]
8 77 ms 77 ms 77 ms if-1-2.tcore2.SQN-SanJose.as6453.net [63.243.205.2]
9 77 ms 77 ms 77 ms if-5-2.tcore2.PDI-PaloAlto.as6453.net [64.86.21.2]
10 78 ms 78 ms 78 ms if-11-3.tcore2.CT8-Chicago.as6453.net [66.198.144.58]
11 79 ms 78 ms 78 ms if-3-2.tcore1.W6C-Montreal.as6453.net [66.198.96.45]
12 81 ms 81 ms 81 ms 66.198.96.50
13 81 ms 81 ms 81 ms 192.34.76.2
14 82 ms 81 ms 81 ms 199.91.189.234
15 81 ms 81 ms 81 ms 199.91.189.36

[whilke]

I do use Google DNS, but that only affects routing when you are dealing with a CDN or similar with multiple IPs servicing the same content.

In this case, the FFXIV world servers have static IPs, with routes determined by advertised BGP tables between core backbone routers and ISP endpoints. I checked the Looking Glass traceroutes, and it seems Telia, Global Crossing, and Sprint always go identical routes through as6453.net to the NA datacenter from my geographical location.

Level3 seems to go through Cogent, but more hops, which is likely why it's not preferred by BGP. This other route which goes through Cogent directly to 192.34.76.2 may be more rarely used for all I know. I'll see what the Looking Glass says regarding latency during peak time.

Worth noting that now that the servers are all down for maintenance and empty of users, the tracert is normal. Definitely suggests S-E may not have purchased enough capacity on the as6453.net backbone to handle the increased active users from their server upgrades these past few weeks:

1 <1 ms <1 ms <1 ms 192.168.1.254
2 1 ms 1 ms <1 ms 001.131-78-65.ftth.swbr.surewest.net [65.78.131.1]
3 1 ms 1 ms 1 ms 172.22.5.13
4 1 ms <1 ms <1 ms 172.21.2.13
5 1 ms <1 ms <1 ms 172.21.0.250
6 4 ms 4 ms 4 ms sjo-bb1-link.telia.net [213.248.88.73]
7 4 ms 4 ms 4 ms tata-ic-140123-sjo-bb1.c.telia.net [213.248.95.130]
8 77 ms 77 ms 77 ms if-1-2.tcore2.SQN-SanJose.as6453.net [63.243.205.2]
9 77 ms 77 ms 77 ms if-5-2.tcore2.PDI-PaloAlto.as6453.net [64.86.21.2]
10 78 ms 78 ms 78 ms if-11-3.tcore2.CT8-Chicago.as6453.net [66.198.144.58]
11 79 ms 78 ms 78 ms if-3-2.tcore1.W6C-Montreal.as6453.net [66.198.96.45]
12 81 ms 81 ms 81 ms 66.198.96.50
13 81 ms 81 ms 81 ms 192.34.76.2
14 82 ms 81 ms 81 ms 199.91.189.234
15 81 ms 81 ms 81 ms 199.91.189.36

That trace doesn't really look very good. You're still jumping up 70ms between two telia hops which are both in the SJO area. That's during non prime time as well...