> On Nov 27, 2017, at 1:18 PM, Jonathan Morton <chromatix99@gmail.com> wrote:
> Here's the difference between "srchost" and "dual-srchost": the latter imposes per-flow fairness on traffic to each host, with a separate queue/AQM per flow like with "flows".  The former only has one queue/AQM per host.
> 
> Analogously for dsthost.
> 
> Then "hosts" mode allocates a separate queue for each host-pair encountered.
> 
> But "triple-isolate" isn't quite analogous to "hosts".  Instead it tries to heuristically behave like either of the dual modes, depending on which one is likely to be on the LAN side of the link.  This allows it to be a reasonable default setting, though the "dual" modes will perform more reliably if chosen correctly.
> 

Thanks, so as I understand it, fairness at the host level for srchost and dual-srchost should be the same, only with dual-srchost there should additionally be fairness among each host’s flows.

If that’s right, I’m confused by this result where there are four clients, each with a separate source IP connecting to the same server IP:

1: tcp_up with 1 flow
2: tcp_up with 12 flows
3: tcp_down with 1 flow
4: tcp_down with 12 flows

With srchost/dsthost it’s fair at the host level (easiest to look at TCP upload/TCP upload sum and TCP download/TCP download sum numbers at the bottom of the page):

http://www.drhleny.cz/bufferbloat/cake/round1/hostiso_eg_cake_src_cake_dst_900mbit/index.html <http://www.drhleny.cz/bufferbloat/cake/round1/hostiso_eg_cake_src_cake_dst_900mbit/index.html>

and with dual-srchost/dual-dsthost it’s not:

http://www.drhleny.cz/bufferbloat/cake/round1/hostiso_eg_cake_dsrc_cake_ddst_900mbit/index.html <http://www.drhleny.cz/bufferbloat/cake/round1/hostiso_eg_cake_dsrc_cake_ddst_900mbit/index.html>