Re: Proposal: Increase QUIC Amplification Limit to 5x

[Matthias Waehlisch <[e-mail]>]

Hi Nick,

  personally, i'm still more in favor of getting certificate sizes 
smaller instead of increasing the amplification limit, since unnecessary 
large certificates are the major problem [1].


cheers
  matthias

[1] https://dx.doi.org/10.1145/3555050.3569123, preprint available on 
https://arxiv.org/pdf/2211.02421

On Tue, 30 Jul 2024, Nick Banks wrote:

> 
> Hello Folks,
> 
>  
> 
> We’ve had this discussion on Slack in the past, and I wanted to bring it here
> to get some additional feedback. As some of you know, I have a project on
> GitHub (microsoft/quicreach) that is a simple ping-like reachability tool for
> QUIC, and I run a periodic action to test the top 5000 hostnames for
> QUIC-reachability and then breaks the handshake down by whether it (a)
> requires multiple round trips, (b) exceeds the specified amplification limit
> or (c) connects in 1-RTT under the limit. It produces this dashboard:
> 
>  
> 
> [IMAGE]
> 
>  
> 
> The main point in sending this email is to focus on the large percentage of
> servers that are ignoring the 3x amplification limit today, and what we should
> do (if anything) about that. I ran a quick experiment (PR) this morning to
> test how the breakdown would look if we had different amplification limits
> (3x, 4x, 5x) and found that if we used a 5x limit we would find ourselves in a
> place where most servers are now under the limit.
> 
>  
> 
> [IMAGE]
> 
>  
> 
> So, my ask to the group is if we should more officially bless a 5x limit as
> ‘Ok’ for servers to use. This would more impact those servers that currently
> take multiple round trips because they are correctly enforcing the 3x limit on
> themselves, resulting in longer handshake times. If we say they can/should
> change their logic from 3x to 5x, then their handshake times will improve, and
> largely things will speed up for clients when using QUIC. Personally, I’d like
> to update MsQuic to use this new limit based on this data, but I wanted to get
> a feel from the group first.
> 
>  
> 
> Thanks,
> 
> - Nick
> 
>  
> 
> Sent from Outlook
> 
> 
> 


-- 
Matthias Waehlisch
.  TU Dresden, Chair of Distributed and Networked Systems
.. https://tu-dresden.de/cs/netd/about/mw