Development discussion of WireGuard
 help / color / mirror / Atom feed
* Embedded x86 Benchmarks v0.0.20180202
@ 2018-02-03 19:17 Lonnie Abelbeck
  0 siblings, 0 replies; only message in thread
From: Lonnie Abelbeck @ 2018-02-03 19:17 UTC (permalink / raw)
  To: WireGuard mailing list

Greetings,

I have a collection of lower-end (embedded) x86 boxes that I use for =
testing with our AstLinux project.

I previously posted WireGuard performance results for v0.0.20171211 ...
Embedded x86 Benchmarks
https://lists.zx2c4.com/pipermail/wireguard/2017-December/002204.html

Results for v0.0.20180202 are below.

Executive summary, v0.0.20180202 ranges from 0.7% slower to 2.9% faster =
than v0.0.20171211.

If nothing else, a good sanity check.

Lonnie

=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=
=3D=3D=3D=3D=3D=3D=3D=3D

All WireGuard endpoints:
Linux 3.16.51
WireGuard 0.0.20180202

Common "Server" endpoint:

VMware VM
x86_64, 2-core, 3.7 GHz Xeon E5-1620
--
vmware-vm ~ # iperf3 -s
--

"Client" endpoints:

Qotom Q190G4N-S07
x86_64, 4-core, 2.0 GHz Celeron J1900
--
[  4]   0.00-30.00  sec  3.03 GBytes   867 Mbits/sec   14             =
sender
[  4]   0.00-30.00  sec  3.03 GBytes   867 Mbits/sec                  =
receiver
--
0.7% slower than v0.0.20171211

Jetway NF9HG-2930
x86_64, 4-core, 1.8 GHz Celeron N2930
--
[  4]   0.00-30.00  sec  3.03 GBytes   867 Mbits/sec   11             =
sender
[  4]   0.00-30.00  sec  3.03 GBytes   866 Mbits/sec                  =
receiver
--
0.7% slower than v0.0.20171211

Lanner FW-7541D
x86_64, 2-core,4-thread, 1.8 GHz Atom D525
--
[  4]   0.00-30.00  sec  2.33 GBytes   668 Mbits/sec    0             =
sender
[  4]   0.00-30.00  sec  2.33 GBytes   667 Mbits/sec                  =
receiver
--
0.1% slower than v0.0.20171211

PC Engines APU2
x86_64, 4-core, 1.0 GHz AMD GX-412TC
--
[  4]   0.00-30.00  sec  1.84 GBytes   526 Mbits/sec    0             =
sender
[  4]   0.00-30.00  sec  1.83 GBytes   525 Mbits/sec                  =
receiver
--
2.9% faster than v0.0.20171211

Lanner LEC-7220-N4
i686, 2-core,4-thread, 1.9 GHz Atom N2800
--
[  4]   0.00-30.00  sec  1.15 GBytes   328 Mbits/sec    0             =
sender
[  4]   0.00-30.00  sec  1.14 GBytes   327 Mbits/sec                  =
receiver
--
0.9% faster than v0.0.20171211

Soekris net5501
i586, 1-core, 0.5 GHz AMD Geode LX
--
[  4]   0.00-30.00  sec  76.9 MBytes  21.5 Mbits/sec    0             =
sender
[  4]   0.00-30.00  sec  76.8 MBytes  21.5 Mbits/sec                  =
receiver
--
0.5% faster than v0.0.20171211

=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=
=3D=3D=3D=3D=3D=3D=3D=3D=

^ permalink raw reply	[flat|nested] only message in thread

only message in thread, other threads:[~2018-02-03 19:11 UTC | newest]

Thread overview: (only message) (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2018-02-03 19:17 Embedded x86 Benchmarks v0.0.20180202 Lonnie Abelbeck

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).