[ovs-dev] [openflow-discuss] Quick experimenting with OVS on OpenWrt

[Alison Chan]

2014-04-08 18:51 GMT-04:00 Nicholas Bastin <nick.bastin at gmail.com>:
> There are two variants of this device - one with an AR9132 (boring white
> case), and one with a QCA9558 (blue spaceship-looking thing).  The AR9132
> version uses a separate switch ASIC (AR8316 - I have one of these), so your
> max throughput through that device in any software forwarding (regardless of
> CPU speed) will be a gigabit.  The 400Mhz ARM in that box is going to limit
> you even further, so 450Mbits seems a bit suspect.
>

We have five of the TL-WR1043ND v1 boring white boxes. They have
Realtek RTL8366RB[0] switch IC if a photo[1] on the openwrt wiki is to
be believed. I don't have my screwdrivers with me to open one up and
confirm, not that my professor would be okay with it ;)
Also, the AR9132 is mips not arm if $(uname -m) is to be believed.

Do you have any suggestions of how to confirm that the 450Mbit/s is
not a false result? I am still a bit disbelieving. I did set-fail-mode
secure and confirmed that no traffic passed with empty flow tables.

[0] http://realtek.info/pdf/rtl8366_8369_datasheet_1-1.pdf
[1] http://wiki.openwrt.org/_media/inbox/tl-wr1043nd_v14.jpg?cache=

Cheers,
-- 
Alison Chan
chan7781 at kettering.edu
SMS: +1 909 278 7753
Fax: +1 206 203 2793