[PATCH v2 2/3] nvme-tcp: support specifying the congestion-control

[Mingbao Sun]

On Mon, 28 Mar 2022 21:33:53 -0700
Jakub Kicinski <kuba at kernel.org> wrote:

> On Tue, 29 Mar 2022 10:48:06 +0800 Mingbao Sun wrote:
> > A server in a data-center with the following 2 NICs:
> > 
> >     - NIC_fron-end, for interacting with clients through WAN
> >       (high latency, ms-level)
> > 
> >     - NIC_back-end, for interacting with NVMe/TCP target through LAN
> >       (low latency, ECN-enabled, ideal for dctcp)
> > 
> > This server interacts with clients (handling requests) via the fron-end
> > network and accesses the NVMe/TCP storage via the back-end network.
> > This is a normal use case, right?  
> 
> Well, if you have clearly separated networks you can set the congestion
> control algorithm per route, right? man ip-route, search congctl.

Cool, many thanks for the education.

I verified this approach, and it did work well.
And I furtherly found the commit
‘net: tcp: add per route congestion control’ which just
addresses the requirement of this scenario (separated network).

So with this approach, the requirements of our use case are
roughly satisfied.

Thanks again ^_^