Co-location deals for Vancouver, BC?

[moolah]

Hi
I'm just seeing if anybody knows of any great deals (under $200/month) on co-locating in Vancouver, BC?


Much appreciated.

[James]

The only host i can think of in that area is Superb Internet www.superb.net You can find their colo prices:

http://services.superb.net/colo/colo.php

[thecoolh]

heh... superb doesn't have a datacenter in vancouver... vancouver's colo rates SUCK... (i live here...)

note: do NOT go near smarttnet... they killed my business, those bastards... lol... would go into the colo room and just pop the reset with no justification... just to get me to leave instead of them illegally kicking me off (violating their contract) because I used too much b/w... I DO have solid proof of it as well...

[thecoolh]

ohh... if you absolutely require colo in vancouver and can't live without it... try out primus telecom... (www.primus.ca)... about $400 USD/Mbps tho... not sure about rackspace rates... drop me a line (email: netdude@telus.net, icq: 74440896, aim: NetDudeMcAffee) and I can get you in touch with my account manager there and you can talk to her about it... another option is peer1 networks (they are NOT tier-1, lol; www.colobrokers.com)... they are like $5/GB i think... hmm... not sure about colo space rates on 'em neither... option #3 would be NetNation's datacenter in harbour center... (http://www.netnation.com/products/colocation.cfm)

lol... ya know what just hit me? get a regular DSL connection from aicom.com and pop a server onto that... hehehe... :) has to be aicom.com because the b/w from all the other providers is charged for... heh... i'm lucky, i was one of telus' first xDSL customers, i don't get charged for b/w usage... heh... :)

[moolah]

I appreciate your rapid responses!

AFter conducting some research last night, I've found DOllarhost to charge the best rates for my bandwidth/traffic needs.

Superb Internet will be the best provider for co-location once our bandwidth/traffic needs increase substantially.

As for Telus, they didn't have any prices quoted on their site. I'll talk to them Mon and see what they have to offer.

[thecoolh]

o yeah... forgot about them... heh... owned by in2net... very honest business... i used to have a server colocated with them 3 years back... lol... or was it 4... hmm... hehe... i did too much b/w for them to handle at the time... (took their entire 10Mbps feed from metronet down, their own website was slowed to a crawl)... and they gave me a FULL refund... and was very pleasant about the whole issue...

at the time, 10Mbps was a LOT of bandwidth... also... does superb have a vancouver datacenter? i didn't think they had one... hmm... ? their offices are here, but datacenter in washington dc?

[moolah]

thecoolh,
you're right about Superb Internet's datacenter being in Washington DC. The company's based in Vancouver, but sales is probably all they do out of the local office.

I'm still waiting for a response from Telus and Peer 1 Networks, but for now, I'm leaning towards Dollarhost.

So I take it you've been managing servers for quite some time. Where do you currently have your personal servers?

[thecoolh]

i've got a couple dozen servers all over the place... my primary service provider is ardent communications (www.ardentcomm.com) in mclean, va... :) $225/Mbps... unless one gets it thru me... heh... :)

[kinetic]

This doesn't lead thru' CIHost anyway ? Better check their trace route, that should show it all.

Secondly, did you check at one company binary-environments.com, I think they do colo in Vancouver thru Peer1. Check them out.

Superb Internet, is having a colo facility in Vancouver also. Better check with them. Infact they are having another facility at Washington,DC. where they have day-time & night-time support. I found they have support available in all time zones to support customers like you.

[kinetic]

Hey moolah, how does it matter, if the servers are in US ? Colo support is always available, round-the-clock on ticket based system working from your browser. Such ticket based support systems are far more advanced and faster, as they get your query directly to the technician at the NOC/DataCenter.

Also, better tracert and see the packet route yourself.

I found the follwing from New Delhi, India:

02/05/02 15:16:34 Fast traceroute www.ardentcomm.com
Trace www.ardentcomm.com (63.217.26.47) ...
1 192.168.1.1 1ms 1ms 1ms TTL: 0 (MAIL ok)
2 192.168.1.254 2ms 16ms 2ms TTL: 0 (No rDNS)
3 203.122.42.1 111ms 97ms 99ms TTL: 0 (No rDNS)
4 203.122.61.13 95ms 96ms 98ms TTL: 0 (No rDNS)
5 203.122.61.17 119ms 105ms 97ms TTL: 0 (No rDNS)
6 203.200.64.225 106ms 118ms 97ms TTL: 0 (No rDNS)
7 203.200.87.15 114ms 126ms 112ms TTL: 0 (No rDNS)
8 202.54.2.241 117ms 122ms 114ms TTL: 0 (vsb-delhi-stm1.Bbone.vsnl.net.in bogus rDNS: host not found [authoritative])
9 209.58.17.5 591ms 614ms * TTL: 0 (if-4-0.bb7.NewYork.Teleglobe.net ok)
10 207.45.220.209 629ms 639ms * TTL: 0 (if-3-2.core2.NewYork.Teleglobe.net ok)
11 207.45.223.177 659ms 615ms 623ms TTL: 0 (if-3-0.core1.NewYork.Teleglobe.net ok)
12 207.45.196.166 620ms 647ms 656ms TTL: 0 (ix-2-0.core1.NewYork.Teleglobe.net ok)
13 205.171.230.22 658ms 635ms 650ms TTL: 0 (jfk-core-02.inet.qwest.net ok)
14 205.171.230.1 * 618ms 631ms TTL: 0 (jfk-core-01.inet.qwest.net ok)
15 205.171.5.235 650ms 628ms 636ms TTL: 0 (wdc-core-02.inet.qwest.net ok)
16 205.171.24.6 634ms * * TTL: 0 (wdc-core-03.inet.qwest.net ok)
17 205.171.8.213 654ms 649ms * TTL: 0 (dca-core-03.inet.qwest.net bogus rDNS: host not found [authoritative])
18 205.171.9.86 653ms 748ms 622ms TTL: 0 (dca-edge-01.inet.qwest.net bogus rDNS: host not found [authoritative])
19 63.216.1.54 * 648ms 612ms TTL: 0 (pos2-0.core2.wdc.cais.net ok)
20 63.216.1.13 641ms 622ms 632ms TTL: 0 (pos3-0.core1.wdc.cais.net ok)
21 63.216.0.70 * 651ms 641ms TTL: 0 (pos5-1.core1.mcl.cais.net ok)
22 63.216.0.34 648ms * 611ms TTL: 0 (pos5-0.colo1.mcl.cais.net ok)
23 63.217.26.47 * 773ms 656ms TTL: 41 (www.ardentcomm.com ok)
24 63.217.26.47 677ms TTL: 41 (www.ardentcomm.com ok)

Hey CoolH, better check this. Why are so many hops there ? might be moving traffic their (www.ardentcomm.com) thru Public Exchanges, all clogged and slow

02/05/02 15:19:15 Fast traceroute www.superb.net
Trace www.superb.net (207.228.225.56) ...
1 192.168.1.1 1ms 1ms 1ms TTL: 0 (MAIL ok)
2 192.168.1.254 2ms 2ms 2ms TTL: 0 (No rDNS)
3 203.122.42.1 239ms 96ms 131ms TTL: 0 (No rDNS)
4 203.122.61.13 98ms 97ms 98ms TTL: 0 (No rDNS)
5 203.122.61.17 189ms 109ms 436ms TTL: 0 (No rDNS)
6 203.200.64.225 104ms 103ms 121ms TTL: 0 (No rDNS)
7 203.200.87.15 228ms 188ms 138ms TTL: 0 (No rDNS)
8 202.54.2.241 116ms 144ms 505ms TTL: 0 (vsb-delhi-stm1.Bbone.vsnl.net.in bogus rDNS: host not found [authoritative])
9 64.211.60.249 * * 1021ms TTL: 0 (so-2-3-3.ar2.NYC2.gblx.net bogus rDNS: host not found [authoritative])
10 208.48.234.205 655ms 1126ms 769ms TTL: 0 (pos5-0-622M.cr2.NYC2.gblx.net bogus rDNS: host not found [authoritative])
11 208.178.174.82 * 1065ms 695ms TTL: 0 (pos0-0-622M.cr2.WDC2.gblx.net ok)
12 206.132.113.134 659ms 1301ms 640ms TTL: 0 (pos1-0-0-155M.ar1.WDC2.gblx.net fraudulent rDNS)
13 204.246.205.50 625ms 1605ms 648ms TTL: 0 (s3-0.core2.dca1.hopone.net ok)
14 207.228.224.7 644ms 1363ms 635ms TTL: 0 (rsm1.dist2.dca1.hopone.net ok)
15 207.228.225.56 657ms 1437ms 669ms TTL:238 (www.superb.net ok)

This Superb seems faster, just see their hops from VSNL, Bombay to them, just 5.

02/05/02 15:19:30 Fast traceroute www.peer1.net
Trace www.peer1.net (64.69.87.230) ...
1 192.168.1.1 1ms 1ms 1ms TTL: 0 (MAIL ok)
2 192.168.1.254 2ms 2ms 2ms TTL: 0 (No rDNS)
3 203.122.42.1 101ms 96ms 278ms TTL: 0 (No rDNS)
4 203.122.61.13 235ms 142ms 139ms TTL: 0 (No rDNS)
5 203.122.61.17 301ms 104ms 108ms TTL: 0 (No rDNS)
6 203.200.64.225 190ms 137ms 105ms TTL: 0 (No rDNS)
7 203.200.87.15 109ms 137ms 336ms TTL: 0 (No rDNS)
8 202.54.2.241 356ms 117ms 368ms TTL: 0 (vsb-delhi-stm1.Bbone.vsnl.net.in bogus rDNS: host not found [authoritative])
9 64.211.60.249 * 625ms * TTL: 0 (so-2-3-3.ar2.NYC2.gblx.net bogus rDNS: host not found [authoritative])
10 208.48.234.201 826ms 704ms * TTL: 0 (pos4-0-622M.cr1.NYC2.gblx.net bogus rDNS: host not found [authoritative])
11 64.212.98.218 714ms * 724ms TTL: 0 (pos3-0-622M.cr1.SEA1.gblx.net ok)
12 206.132.41.126 * * 735ms TTL: 0 (No rDNS)
13 208.50.86.18 748ms * 908ms TTL: 0 (Peer1net.ge-3-1-0.ar2.SEA1.gblx.net bogus rDNS: host not found [authoritative])
14 64.69.67.129 721ms 718ms 721ms TTL: 0 (POS0-2.van-gsr-a.peer1.net bogus rDNS: host not found [authoritative])
15 64.69.87.230 * 722ms 736ms TTL:104 (peer1.net ok)
16 64.69.87.230 746ms TTL:104 (peer1.net ok)

Peer1.net Do they have any better connectivity Except Vancouver, that too, its not multiple redundant, check out their Internal network. Can't offer you redundancy beyond backbones.

And CoolH, do Ardent have their own technical people in this datacenter to look after the servers ?

hehe

[thecoolh]

lol... ticket-based system via browser is bull*** compared to 24/7 1-800 telephone support... as for techs in noc to look after servers, yes, ardent has them on staff IN-THE-DATACENTER (not on-call) 24/7...

about that ardentcomm traceroute... what the hell are you talking about? that specific traceroute doesn't show ANY public exchange points! which IP on there is a public exchange point? superb vs ardent off by a few ms on the two... that few ms is not bad considering the extra hops... but ardent is tier-1 and superb is tier-3... makes a world of difference...

also... look at the superb traceroute and the ardentcomm traceroute... it stays steadily near the average levels... while the superb traceroute max bursts above quite a bit...

but then again... these are traceroutes from india... doesn't really help evaluate a service provider that well... heh... fyi... ardent does NOT buy transit from glbx.net (global crossing, currently going bankrupt... heh)... but they do peer with them... heh... :) direct private peer, not via MAE-EAST... heh...

about the peer1 networks... they are a decent 'local' service provider... feeds to 3-4 different transit providers... similar to superb's setup... but they're on the west coast vs superb's east coast datacenter... which would explain why their ping times are worse than the other two providers... ur transit from india comes in via the east coast... ! :P both ardent and superb = east coast.
---------------------------------------------------------------------------------
traceroute to www.globalcrossing.com (209.130.223.50), 30 hops max, 40 byte packets
1 ge3-0.colo1.mcl.cais.net (205.252.89.1) 0.366 ms 0.569 ms 0.378 ms
2 pos5-2.core2.mcl.cais.net (63.216.0.37) 0.474 ms 1.098 ms 0.403 ms
3 at0-0-0-247.br1.WDC2.gblx.net (208.49.231.5) 1.306 ms 1.938 ms 1.758 ms
4 pos2-0-155M.cr1.WDC2.gblx.net (208.178.174.53) 1.281 ms 1.344 ms 1.143 ms
5 pos2-0-2488m.cr1.CHI1.gblx.net (208.49.59.250) 32.587 ms 32.471 ms 31.743 ms
6 pos0-0-0-155M.ar1.CHI1.gblx.net (206.132.118.82) 32.161 ms 31.952 ms 33.147 ms
7 FrontierOnline.s4-0-1.ar1.CHI1.gblx.net (206.165.81.202) 36.725 ms 36.494 ms 37.185 ms
...

[thecoolh]

whats that about cihost?

o yeah... n since ur advocating superb so much... just thought i'd let y'all know... ardent is a hella lot bigger... heh... as a business and as a network...

[kinetic]

Hey CoolH,

No advocating here. moolah asked for a vancouver colo, and i happened to know about their support on colo for us in India too. This far away, on the other side of the globe, we don't need 1-800 calls. Web based ticket support is what works faster for us, and should work for anybody who's truly international, not focussing only US & Canada.

We have talked to Superb's colo customers here in india and they regularly get ticket support, day-time here & night-time with you guys. So, a ticket-based support is best as they have night time technicians near the servers, and believe me, the support is faster, if not instant its immediate.

Better still they have technical guys here in India whom we can call at any point in time, we found they receive calls from US, Europe, Australia and Asia Pacific for all night time queries as well.

Not really advocating anyone, as i don't understand that, you being an expert, tell me how is Superb tier-3 ? Do ardent own the complete US backbones ? do they not take backbones from UUNET, Sprint, AT&T, Level3, TeleGlobe, GlobalCrossing ? I mean do they have their own cable network across US and is that cable backbone network better than all these providers ?

As for trace routes, you are correct. For us, some packets travel directly thru' NY and some go all the way thru' Singapore to East Coast.

I think you can guide us all better.

Cheer up

[thecoolh]

yes, ardent does own a complete nationwide backbone... they peer directly with uunet, sprint, level3, etc/etc/etc... its not a matter if being 'better'... its a matter of connectivity... redundancy and accessibility to end users... and the way you're putting it its as if they have only a single fiber going across the country... ardent does have an extensive 'web' across the US...

as for the ticket based system, i see your point... ardent is actually in the process of implementing that for the benefit of international clients i think... heh...

[kinetic]

Hey CoolHost,

How's that Superb is Tier-3 as you say so.

I understand that your Ardent is tier-1, but what percent of internet traffic is on their backbone ?.....hehe

and I find in their trace route that their backbones also have Packet-Over-SONET, the technology that you keep promoting is also Packet over------ technology......no real difference there....;)

Cheers.
Kinetic

[thecoolh]

heh, are you contesting that Superb is tier-1?

as for how much of the 'net runs of ardent's backbone, does that really matter? the fact that it has direct access onto most of the other tier-1 networks around is the whole point of being tier-1. Which results in more direct routes to most of the other providers around. :)

As far as the difference between Packet-Over-SONET vs ATM, the performance difference is about 30%. That is due to the lower overhead of the packets themselves, but with this extra performance you lose the integrated packet shaping abilities which are a part of ATM... But with most things regarding the 'net, the integrated packet shaping isn't needed as it can be done easily enough at the router level.

[Mr Chunder]

These guys come out in the colo search on this site:

http://www.hbhosting.com/index1.html
http://www.netnation.com/
http://www.actadivina.com/

[thecoolh]

ahh... i see the relevance... :D but... hbhosting has their datacenter in texas while netnation and actadivina are in vancouver (both get their feeds via peer1.net)