Re: multiplexing -- don't do it

[Salvatore Loreto <[e-mail]>]

I concur with Robert that solving a transport problem is decidedly not 
in scope of the WG.

however I can not resist ...


On 4/8/12 12:43 AM, Adrien W. de Croy wrote:
> also... UDP is very problematic for DoS, since there's no established 
> connection, and therefore no verification of source.
right, but you can use DTLS for this
> Try defending yourself against a DNS flood with spoofed source and 
> you'll see what I mean.  The work you need to do to respond to an HTTP 
> request is typically a lot more than you need to do for DNS, so the 
> impact would be worse.
> So IMO it's completely unsuitable as a transport for HTTP over the 
> internet.
but you can use SCTP over DTLS over UDP can provide an interesting 
solution to investigate from a *research prospective*
SCTP in the userland provide you with natively MUX and congestion 
control, and the userland approach solve the fact that not all the OSs 
support SCTP;
DTLS provides you with security
UDP: encapsulation of SCTP over UDP allows also to pass through NAT

/Sal

-- 
Salvatore Loreto, PhD
www.sloreto.com