Thank you for your inputs - @Jordan, @Tom, @Christian On 1/30/2020 9:07 PM, Tom Smyth wrote: > Livio are you running iperf on the apu ? > The apu doesnt have much cpu to generate packets from iperf... > Forwarding perf should about 450m on an apu c2 with pf enabled and about > 850m-900m with pf disabled > That issting with iperf through the apu2c2 with decent professional laptops > with iperf generated traffic and measured on these laptops
@Tom: Yes, I am running iperf (server) on the APU. When I run iperf (server) on the notebook I get the following results: apu# iperf3 -c 192.168.20.40 Connecting to host 192.168.20.40, port 5201 [ 5] local 192.168.20.28 port 17990 connected to 192.168.20.40 port 5201 [ ID] Interval Transfer Bitrate [ 5] 0.00-1.00 sec 81.5 MBytes 680 Mbits/sec [ 5] 1.00-2.00 sec 81.1 MBytes 681 Mbits/sec [ 5] 2.00-3.01 sec 82.5 MBytes 685 Mbits/sec On 1/30/2020 11:29 PM, Christian Weisgerber wrote: > I vaguely remember a thread somewhere that concluded that one of > these network benchmark tools degenerated into a benchmark of > gettimeofday(2), which apparently is very cheap on Linux and not > cheap on OpenBSD. So you end up measuring the performance of this > system call. > > I don't remember whether it was iperf... You are probably right. I will now have to test the throughput with tcpbench as suggested by Jordan. On 1/31/2020 2:04 AM, Jordan Geoghegan wrote: > That sounds about right. I vaguely remember reading a thread about iperf on > misc some time in the past year mentioning that. > While OpenBSD obviously doesn't have the same network performance as Linux or > FreeBSD, as work continues on unlocking more of the kernel, things will > continue to get better. I think bluhm@ regularly runs some automated > benchmarks that show that OpenBSD maxes out at around 4-5 Gbit / second > throughput. Thank you, I will try to run some tests using tcpbench and send an update. The forwarding performance is the relevant part.
