Which datacenter should i choose?
I want backups, and so does he, but i don't want to pay another $5 a month for it. So, i suggested that we backup to each other. Once we came to that, we decided different datacenters would probably be best. He's in Newark, i'd like to choose the one with the least latency. I'm guessing it would be Dallas or Atlanta. But a real test might be in order.
We both live in Michigan.
Any ideas?
And, does inter-datacenter traffic go against by limit?
11 Replies
@chacham:
And, does inter-datacenter traffic go against by limit? Why wouldn't it? All EXTERNAL traffic counts.
Any comments as to which datacenter?
http://www.linode.com/speedtest/
@chacham:
I want backups, and so does he, but i don't want to pay another $5 a month for it. So, i suggested that we backup to each other. Once we came to that, we decided different datacenters would probably be best. He's in Newark, i'd like to choose the one with the least latency. I'm guessing it would be Dallas or Atlanta. But a real test might be in order.
Stay away from Fremont – it's been having continuing backup issues for the past couple of weeks. I haven't had a successful backup in five days.
Of course, if you're doing your own backup, then heck, you probably are better off.
thanks,
bruce
Just my two cents.
He tested the downloads this morning from a Linode in NJ. Here are the results:
London: 63 Kps
Atlanta: 15000 Kps
Dallas: 7500 Kps
Fremont: 3500 Kps
Anyone else want to try? The Atlanta numbers seem almost too good to be true.
–
He ran the Atlanta one a few times to verify. He figures they're high on Coke.
I'd love to see numbers from someone else in NJ as well. (Or even Atlanta, for the NJ download.)
Dallas
–2011-06-20 12:27:10--
Resolving newark1.linode.com… 207.192.68.6
Connecting to newark1.linode.com|207.192.68.6|:80… connected.
HTTP request sent, awaiting response… 200 OK
Length: 104857600 (100M) [application/octet-stream]
Saving to: `100MB-newark.bin'
100% 104,857,600 6.26M/s in 15s
2011-06-20 12:27:25 (6.87 MB/s) - `100MB-newark.bin' saved [104857600/104857600]
London
–2011-06-20 12:28:35--
Resolving newark1.linode.com… 207.192.68.6
Connecting to newark1.linode.com|207.192.68.6|:80… connected.
HTTP request sent, awaiting response… 200 OK
Length: 104857600 (100M) [application/octet-stream]
Saving to: `100MB-newark.bin'
100% 104,857,600 4.12M/s in 25s
2011-06-20 12:29:00 (4.02 MB/s) - `100MB-newark.bin' saved [104857600/104857600]
I'd say your london test was wayyyy off.
London
–2011-06-20 08:52:29--
Resolving london1.linode.com… 109.74.207.9
Connecting to london1.linode.com|109.74.207.9|:80… connected.
HTTP request sent, awaiting response… 200 OK
Length: 104857600 (100M) [application/octet-stream]
Saving to: `100MB-london.bin'
100%[======================================>] 104,857,600 74.6K/s in 20m 50s
2011-06-20 09:13:19 (81.9 KB/s) - `100MB-london.bin' saved [104857600/104857600]
Dallas
Resolving dallas1.linode.com… 69.164.200.100
Connecting to dallas1.linode.com|69.164.200.100|:80… connected.
HTTP request sent, awaiting response… 200 OK
Length: 104857600 (100M) [application/octet-stream]
Saving to: `100MB-dallas.bin'
100%[======================================>] 104,857,600 6.21M/s in 14s
2011-06-20 09:17:03 (6.92 MB/s) - `100MB-dallas.bin' saved [104857600/104857600]
I would say the London results are not that off
I would guess it depends on routing…
–2011-06-20 13:25:51--
Resolving london1.linode.com… 109.74.207.9
Connecting to london1.linode.com|109.74.207.9|:80… connected.
HTTP request sent, awaiting response… 200 OK
Length: 104857600 (100M) [application/octet-stream]
Saving to: `100MB-london.bin'
100% 104,857,600 2.39M/s in 36s
2011-06-20 13:26:26 (2.81 MB/s) - `100MB-london.bin' saved [104857600/104857600]
I wonder why downloading london from newark is so pants for you guys but else where it's fine :/