Re: [Cerowrt-devel] performance numbers from WRT1200AC (Re: Latest build test - new sqm-scripts seem to work; "cake overhead 40" didn't)

[Dave Taht <dave.taht@gmail.com>]

On Tue, Jun 23, 2015 at 5:55 AM, Mikael Abrahamsson <swmike@swm.pp.se> wrote:
> On Tue, 23 Jun 2015, Sebastian Moeller wrote:
>
>>         Most likely not. Check http://wiki.openwrt.org/doc/howto/sqm .
>> Rich published a great set of instructions for setting up sqm-scripts under
>> openwrt proper.
>
>
> I tried it on Linksys WRT1200AC with OpenWrt CC RC2. I configured sqm to
> have 800 megabit/s each direction, and ran iperf3 over IPv4 with NAT44 from
> Linux box behind WRT1200AC to an OSX macbook connected on a switch on the
> same L2 subnet as the WAN port.
>
> Linux <->WRT1200AC<->switch<->OSX
>
> I get 765 megabit/s of throughput using single session, at sirq load of
> around 25%. If I lower the mss to 300 (to generate higher pps) I get around
> 560 megabit/s of throughput at 50% sirq. With 10 parallel TCP sessions, I
> get about the same. At MSS of 200 bytes, I get 400 megabit/s at 70% sirq.
>
> If I turn off SQM completely, I get 600 megabit/s at 200 byte MSS single
> session at 80% sirq and 930 megabit/s at 26% sirq with default MSS.

Yer missing the more important figure. What is the induced latency in
all these cases? With the system being software limited, I would
imagine that other oddities arise. Running out of cpu, additional
oddities.

When going at hardware line rate, is fq_codel enabled? Does it ever
engage? (My limited testing showed that lacking BQL, delay accrued big
time in the drivers themselves on this platform)

Good idea on using a reduced MSS size!

I would really like to get to the point where cake ran on this
platform, but thus far we have not managed to get a working build, nor
push cake into mainline openwrt.... my observation is that a lot of
the overhead of sqm comes from tc filters, iptables rules and NAT.

> So if you want high performing device that is OpenWRT compatible and still
> does forwarding using CPU so you can test queuing algorithms, the WRT1200AC
> and WRT1900ACv2 is the best I have been able to find currently (unless you
> go for x86 platform).

yes, x86 is fastest. I have not tried the reduced mss idea on my
rangeley box, I will check that out!

> --
> Mikael Abrahamsson    email: swmike@swm.pp.se
> _______________________________________________
> Cerowrt-devel mailing list
> Cerowrt-devel@lists.bufferbloat.net
> https://lists.bufferbloat.net/listinfo/cerowrt-devel



-- 
Dave Täht
worldwide bufferbloat report:
http://www.dslreports.com/speedtest/results/bufferbloat
And:
What will it take to vastly improve wifi for everyone?
https://plus.google.com/u/0/explore/makewififast