[ih] Better-than-Best Effort

[Brian E Carpenter]

Dave,

On 27-Aug-21 10:39, Dave Crocker via Internet-history wrote:
> Having not followed actual QOS work over the year, my naive brain 
> wandered oddly, today with a thought about a semi-QOS approach.
> 
> The usual view is that it requires complete, end-to-end support. 
> Massive barriers to adoption, at the least.
> 
> I'm thinking that the long-haul infrastructure tends to have enough 
> capacity that it usually isn't the source of latency.  It's the 
> beginning and ending legs that do.
> 
> So what about a scheme that defines and provides QOS in those segments 
> but not the long middle?  Cheaper, more implementable, and might give 
> usefully-better performance.
> 
> Assuming that this idea is new only to me, I'm curious about 
> reactions/history/etc.

That's pretty much the deployment model for diffserv. Apply the
per-hop behaviour that you want locally, and hope that the WAN has
enough capacity. But of course, that is not guaranteed, so everything
is messed up by buffer bloat, so you get "buffering" messages during
your live video. Which is why you need upstream service level agreements,
but you the user have no control over that. Also, your incentives are
not the same as the transit ISPs' incentives, so your local ISP is
caught in the middle, between conflicting incentives.

This is a really complex topic but well known in tsvwg at ietf.org.

   Brian