On Mon, 3 Apr 2023, Ayush Mishra via Bloat wrote:

> ==> I agree too. But I think one of the key challenges here could be when
> the dynamically entering flows are extremely tiny (which I imagine is quite
> common). In those cases, there is a possibility that by the time the
> long-running flow backs off, the congestion it was responding to has
> already ended because the tiny flows have exited the bottleneck (think
> microbursts caused by flows that last 1-2 RTTs). In a perfect world we'd
> like to deal with elephant and mice flows in isolation at the switch, but
> there are likely things we can do from the endpoint too. Maybe some kind of
> a two-phase backoff, with the second phase only kicking in after a period
> of hysteresis to make sure it's responding to persistent congestion and not
> just brief microbursts. This is just off the top of my head, so I'm not
> sure how something like this would play out in the overall dynamics and
> convergence of the algorithm that implements it.

if the backoff is less drastic than currently, I would say that it's reasonable 
to have the elephant back off for congestion caused by a microburst, as such 
bursts are not uncommon, and while the one that triggered the backoff may finish 
before the backoff happens, it's made room for the next one.

David Lang