very low performance tcp/rsh

Alan Somers asomers at freebsd.org
Mon Apr 29 20:08:44 UTC 2019


On Mon, Apr 29, 2019 at 1:57 PM Eugene Grosbein <eugen at grosbein.net> wrote:
>
> 30.04.2019 2:22, Wojciech Puchar wrote:
>
> > create 32GB hole-file
> >
> > dd if=/dev/zero of=test bs=32k seek=1m count=0
> >
> > fetch -o /dev/null over ftp (proftpd) gives 1.4GB/s
> >
> > proftpd consumed 100% of single core. much better.
> >
> > tested with FreeBSD ftpd  - 1.7GB/s
> >
> > seems like freebsd can saturate 10Gb/s ethernet with single core (Intel(R) Xeon(R) CPU E5-2620 v4 @ 2.10GHz)
> >
> >
> > Still where is the most overhead? one memcopy to packet buffers+creating packet headers (send).
> >
> > reception is on separate process taking another core.
> >
> >
> > still - quite good. But compared to filesystem overhead - large.
>
> Do not use file with holes (holes require special processing taking lots of CPU cycles).
> Create ordinary file within RAM disk and use sock ftp daemon for sake of sendfile().

Better yet, if the goal is simply to measure TCP performance, use
benchmarks/netperf.  Don't try to do anything involving files.

-Alan


More information about the freebsd-hackers mailing list