NFS + Infiniband problem

[Andrew Vylegzhanin]

ср, 31 окт. 2018 г. в 5:53, Rick Macklem <rmacklem at uoguelph.ca>:
>
> >
> >net.inet.tcp.cc.algorithm=htcp
> Well, I'm not familiar with the current TCP stack (and, as noted before,
I know
> nothing about InfiniBand). All I can suggest is testing with the default
congestion
> control algorithm. (I always test with the default, which appears to be
newreno.)
> NFS traffic looks very different than a typical use of TCP. Lots of small
TCP
> segments in both directions interspersed with some larger ones (the write
> requests or read replies).

With this TCP settings same server serve NFS requests via 40G Ethernet on
multiply clients with speed via 1G Eth ~ 105/110 MB/sec  write/read.
Of course I'll try to change congestion algorithm, but I don't think that
will help.

Also need to test setup with infniband set from connected mode to datagram
mode.

--
Andrew

>
> rick