[WireGuard] wireguard comparing to fastd - tests

jens jens at viisauksena.de
Wed Aug 17 01:10:44 CEST 2016


On 16.08.2016 14:17, Jason A. Donenfeld wrote:
>> ---------- Forwarded message ----------
>> From: jens <jens at viisauksena.de>
>>
>> thx, thats true ... we were mostly interested in comparing speed in
>> similar setups that we would deploy.
>> But you ' re right - reducing workload from 20 to 12 in cypher make
>> them not directly comparable
> Alternatively, you could pretty easily tweak wireguard to use
> chacha20/12 by changing "for (i = 0; i < 20; i += 2) {" to "for (i =
> 0; i < 12; i += 2) {" in src/crypto/chacha20poly1305.c.
>
> Would be interested in fair side-by-side benchmarks of the two.

here some more tests with fastd directly, i dont see any big difference
between salsa20 and salsa2012, but i big difference in direction ...
which is somehow new for me.

i will not do the different wireguard testings with different ciphers,
as i am still struggling in building valid reproducable kmod..ipk
packages for LEDE

so i think fastd testing is finished by this (which at all only was to
compare to our existing setup) - a small notice : OpenWRT Wiki was wrong
with CPU speed of TPlink841v11 - it is not 560MHz, it is 650Mhz - if
somebody ever want to compare another embedded device

greetz Jens

2016-08-16 06:36:42 +0000 --- Verbose: new session with <foo>
established using method `salsa20+umac'.

TCP
- - - - - - - - - - - - - - - - - - - - - - - - -
[ ID] Interval           Transfer     Bandwidth       Retr
[  4]   0.00-10.00  sec  18.3 MBytes  15.3 Mbits/sec    0             sender
[  4]   0.00-10.00  sec  16.6 MBytes  14.0 Mbits/sec                 
receiver
TCP - other direction
- - - - - - - - - - - - - - - - - - - - - - - - -
[ ID] Interval           Transfer     Bandwidth       Retr
[  4]   0.00-10.00  sec  13.6 MBytes  11.4 Mbits/sec   28             sender
[  4]   0.00-10.00  sec  13.5 MBytes  11.3 Mbits/sec                 
receiver

UDP (pushed 200M udp - somthing up to 30 is fine with many jitter and loss)
[  5] local 192.168.2.23 port 5201 connected to 192.168.1.23 port 44727
[ ID] Interval           Transfer     Bandwidth       Jitter   
Lost/Total Datagrams
[  5]   0.00-1.00   sec  1.15 MBytes  9.63 Mbits/sec  5.144 ms  0/147 (0%) 
[  5]   1.00-2.00   sec  0.00 Bytes  0.00 bits/sec  5.144 ms  0/0 (-nan%) 
[  5]   2.00-3.00   sec  0.00 Bytes  0.00 bits/sec  5.144 ms  0/0 (-nan%) 
... killed/crashed




again
2016-08-16 07:22:41 +0000 --- Verbose: new session with <foo>
established using method `salsa2012+umac'.
TCP direction -> peer - server
- - - - - - - - - - - - - - - - - - - - - - - - -
[ ID] Interval           Transfer     Bandwidth       Retr
[  4]   0.00-10.00  sec  18.4 MBytes  15.4 Mbits/sec   12             sender
[  4]   0.00-10.00  sec  16.7 MBytes  14.0 Mbits/sec                 
receiver
TCP direction -> server - peer
- - - - - - - - - - - - - - - - - - - - - - - - -
[ ID] Interval           Transfer     Bandwidth       Retr
[  5]   0.00-10.04  sec  13.8 MBytes  11.5 Mbits/sec   25             sender
[  5]   0.00-10.04  sec  13.7 MBytes  11.4 Mbits/sec                 
receiver

-- 
make the world nicer, please use PGP encryption

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.zx2c4.com/pipermail/wireguard/attachments/20160817/c15d426d/attachment.html>


More information about the WireGuard mailing list