Maybe we can consider this topic (anycast in content download and DNS) in 
another angle. For content download, the focus is throughput, and the freezing 
time caused by the anycast interruption is OK since the freezing time (which is 
usually several or tens or hundreds of milliseconds) can be omitted compared 
with the whole content download time which may be several or tens of minutes. 
While for DNS, the focus is the latency but not throughput, and the freezing 
time cannot be omitted since one normal DNS resolution latency is usually one 
RTT. So DNS is different from other applications like content download in the 
traffic pattern and QoS (Quality of Service) metric.
 


Guangqing Deng
CNNIC 
 
From: Tony Finch
Date: 2015-04-28 19:34
To: Phillip Hallam-Baker
CC: Warren Kumari; [email protected]; Christian Huitema; Paul Hoffman
Subject: Re: [dns-privacy] DPRIVE over UDP or TCP
Phillip Hallam-Baker <[email protected]> wrote:
> On Tue, Apr 28, 2015 at 5:04 AM, Tony Finch <[email protected]> wrote:
> > Phillip Hallam-Baker <[email protected]> wrote:
> >>
> >> Having it work for content and DNS are two different things. The
> >> routing tables only need to be constant for a few minutes to support
> >> TCP content download. For DNS to be viable they have to be stable much
> >> longer.
> >
> > Why?
>
> The byterange extensions in http mean that it is possible to resume a
> session interrupted part way through if it is static content.
 
Most HTTP interactions aren't resumable in this way, so this observation
does not suggest to me that HTTP needs less routing stability than DNS.
 
> If the anycast changes then you are going to have to timeout and resume.
 
This is also true for HTTP. I still don't see why DNS needs more routing
stability than HTTP.
 
Tony.
-- 
f.anthony.n.finch  <[email protected]>  http://dotat.at/
Rockall, Malin, Hebrides, Bailey: West or northwest, becoming cyclonic 5 to 7,
occasionally gale 8, except in Hebrides. Rough or very rough, occasionally
high at first in Bailey. Rain or showers. Good, occasionally moderate.
 
_______________________________________________
dns-privacy mailing list
[email protected]
https://www.ietf.org/mailman/listinfo/dns-privacy
_______________________________________________
dns-privacy mailing list
[email protected]
https://www.ietf.org/mailman/listinfo/dns-privacy

Reply via email to