DNS vendor performance
I came across this post about DNS performance on Hacker News yesterday, which was interesting because I’d been conducting similar experiments for StyleFeeder. Our site is fast and has scaled well, but I’m always on the lookout to shave off a few milliseconds here and there from our requests. I was looking at our Pingdom reports for DNS recently, and decided to run a few comparisons.
We currently use DynDNS for DNS service. We switched to them probably two years ago after I decided that UltraDNS was stupidly expensive for what they were giving us. In the end, I managed to cancel our UltraDNS service over the phone by providing publicly available information, which was awesome as their standard contract at the time ran on a yearly basis and they ended up screwing me for another year of service; the whole thing left a very bad taste in my mouth. Although, UltraDNS runs some TLDs and big sites, so they obviously have some redeeming qualities. Pricing isn’t one of them. DynDNS was costing us 60x less at the time.
I’ll use the same format as JohnPhilipGreen’s original post to present my numbers. One difference is that my figures were generated using 30 days worth of data, not 3 days as John used.
| DNS Server | Test site | Response time | Standard Deviation |
| ns1.mydyndns.org | www.stylefeeder.com | 112ms | 45ms |
| ns1.contegix.com | www.stylefeeder.net | 112ms | 55ms |
| ns1.dnsmadeeasy.com | www.hubspot.com | 127ms | 97ms |
John basically disqualified DNSMadeEasy as part of his tests because the results were so erratic. I haven’t noticed anything that warrants disqualification like that, but they’re clearly the slowest of the three that I was monitoring.
Some notes: the HubSpot folks are friends of ours, so I figured they wouldn’t mind if I put monitors on their service. Contegix (by far the best managed hosting company I’ve ever worked with) runs our non-cloud infrastructure for us; they offer DNS service at no additional charge to hosting clients and we use it for one of our auxiliary domains, primarily for legacy reasons at this point.
Dynect was the big winner in John’s shootout, which is interesting, given that DynDNS and Dynect are part of the same overall parent company. They currently charge USD $27.50 per year, a far cry from what UltraDNS charges and – I’m guessing – substantially less than what Dynect charges. Is it worth it? For most small sites, probably not… for StyleFeeder, it may very well be worth switching to a faster provider, especially to gain a >50% reduction in response time depending on the cost (John was seeing 42ms as an average response time).
John, maybe you can republish your numbers after 30 days to see how things stand at that point?
Useful data! Thanks for sharing ;) And we don’t mind being part of your benchmark at all.
BTW, the link to HubSpot is broken, missing “.com” at the end of the href ;)
Not anymore! Thanks for letting me know.
Cool comparison, I love to see objective measures on DNS performance. I expect a lot more of this in the future which is great for all of the service providers to see where to improve. Users’ tolerance for latency is a pretty logarithmic which is going to push us even faster. Like John’s report, it’s the Pingdom DNS test? Where are they testing from and how are they averaging?
The Dynect Platform and DynDNS.com are indeed run by the same company but they are really different services. There’s some truth on value discussion for Dynect Platform v. DynDNS/Custom DNS for smaller sites. In the end we think we have both covered.
One thing not highlighted on the Pingdom report is overall availability of DNS services. DynDNS/Custom DNS has been running since 2001 with no outages. Since then, we’ve picked some pretty cool sites, a number of TLDs, and built out a new network with the TLDS/enterprise customers on the Dynect Platform.
Yes, the probes that are measuring performance here are using Pingdom. The locations that Pingdom use are on their site which today are listed as:
1 Vasteras, Sweden s15.pingdom.com 212.247.189.113
2 Herndon, VA s403.pingdom.com 207.97.207.200
3 Dallas 4, TX s402.pingdom.com 67.192.120.134
4 London, UK s401.pingdom.com 78.136.27.223
5 Montreal, Canada s34.pingdom.com 67.205.112.79
6 Houston 2, TX, USA s33.pingdom.com 174.133.253.162
7 Stockholm, Sweden s238.pingdom.com 83.140.19.38
8 Reading, England s31.pingdom.com 89.151.110.152
9 Dallas 3, TX, USA s25.pingdom.com 209.62.54.130
10 Berkeley, CA, USA s16.pingdom.com 72.51.41.47
11 Paris, France s404.pingdom.com 62.193.228.125
Thanks for a helpful report!
It would be interesting to see this study expanded to include some of the registrars like GoDaddy that offer DNS.
It would also be interesting to test eNOM as well.
That’s where HubSpot (and many related sites) were hosting their DNS for a while, before we switched to DNSMadeEasy for some of it.
You can use http://OnStartups.com for your test, if you’d like.
[…] we noticed some room to improve our DNS numbers, which I mentioned in a post last April. Shortly thereafter, I met up with Jeremy Hitchcock from […]
I’m looking at DNS providers to setup failover between two data centres, so very interesting post. Thanks!
My Pingdom trial account is now busily running a DNS 1 minute test for our current rackspace NDS server. I’ll give it a few days and might come back here with the numbers.
BTW – Dynect and Dnsmadeeasy are the only ones mentioned here to employ the anycast technology. I think that should count in their favour?
Even though DynDNS is tun by the same company as Dynect, it is a lower feature service and does not use anycast as far as I understand.
DNS Made Easy is really starting to improve network performance and really starting to kill the performance charts. Unbelievable that they can do this for price that they offer their services for. In recent pingdom results (since this is what everyone is using here) they are on top. So please update your results on this website as this is misleading.
Any chance on running these stats / comparisons again? Data looks to be dated over a year ago. I wonder if some companies have been able to keep their advantage or if the others have caught up.
GoDaddy seems to offer free IP Anycast DNS services now. How do they compare with these high priced and low prices DNS providers?
If interested I’ll run a test against both dynect and dnsmadeeasy servers for the domains in the table.
The http://www.stylefeeder.net domain is no longer resolved, so only the two remaining sites will be tested.
[…] back in April 2009, I read two articles about using Pingdom to test hosted DNS services. Pingdom has a limited free account, includes […]