Re: shim-aware transports

[Iljitsch van Beijnum <iljitsch@muada.com>]

On 16-aug-2005, at 9:43, Pasi Sarolahti wrote:

> I think it would be good also to have some kind of signal from shim to
> TCP to indicate when path (address) has changed from quite early on
> (somewhere between "initially" and "next step").

Yes, early on for the shim. Whether implementers want to change TCP  
immediately is up to them.

> In response TCP should
> re-initialize its congestion control and RTT parameters, and probe the
> new path capacity in slow-start. The new path likely has different
> properties, so slow-start is the correct way of re-probing the current
> path capacity.

I agree. However, at least one person argued that slow start was  
unnecessary/undesirable here.

> (TCP would need to maintain also per address-path cwnd and  
> ssthresh, and
> possibly other internal congestion control-related parameters, in
> addition to RTT and friends)

Yes.

> This seems a pretty significant change to TCP.

Right. But it would allow TCP to use the best path, or even load  
balance across multiple paths. Whether this is worth it is hard to  
tell at this point.

Iljitsch