Re: [dnsoverhttp] Survey of DNS over HTTP

And the update is work in progress? I hope so!

I think the document could usefully address the delay/cost thing you note.

Its worth quantifying. And, as we explore DTLS and different encodings
over HTTP/S, we're going to want a test base to make comparisons over.
I'd suggest a small dns label test set which can be used to verify
additional delay from a prepped cache (ie, remove the onward DNS
variant delay, and fetch solely from an entity which holds the
responses being sought)

Keep-alive? If the session is held open, the TCP establishment costs
are amortized. So, for comparison of DNS over TCP, DNS over HTTP may
have lower aggregate overhead in network protocol terms (obviously the
repeated encodings in the proto layers make this moot)

Nice doc.

_G

On Wed, Sep 14, 2016 at 11:04 PM, Shane Kerr <shane@time-travellers.org> wrote:
> Hello,
>
> I thought that people interested in this topic might find a draft
> written by my colleagues and myself useful. It is a survey of various
> DNS over HTTP techniques:
>
> https://datatracker.ietf.org/doc/draft-shane-review-dns-over-http/
>
> It needs to be updated, to reference Paul Hoffman's recent update to
> his DNS JSON draft if nothing else:
>
> https://tools.ietf.org/html/draft-hoffman-dns-in-json/
>
> Cheers,
>
> --
> Shane
>
> _______________________________________________
> dnsoverhttp mailing list
> dnsoverhttp@ietf.org
> https://www.ietf.org/mailman/listinfo/dnsoverhttp
>