Re: [NNagain] FCC NOI due dec 1 on broadband speed standards

[rjmcmahon <rjmcmahon@rjmcmahon.com>]

iperf 2 supports two features relevant here - one for a live video 
source (isochronous) and the other for DASH like protocols (--burst-size 
& --burst-period.) Note: One can select the CCA to influence pacing or 
not on bursts. These tests are under utilized and not well known. Also, 
don't forget --trip-times on the client

--isochronous[=fps:mean,stdev]
send isochronous traffic with frequency frames per second and load 
defined by mean and standard deviation using a log normal distribution, 
defaults to 60:20m,0. (Note: Here the suffixes indicate bytes/sec or 
bits/sec per use of uppercase or lowercase, respectively. Also the p 
suffix is supported to set the burst size in packets, e.g. 
isochronous=2:25p will send two 25 packet bursts every second, or one 25 
packet burst every 0.5 seconds.)

--burst-period n
Set the burst period in seconds. Defaults to one second. (Note: assumed 
use case is low duty cycle traffic bursts)
--burst-size n
Set the burst size in bytes. Defaults to 1M if no value is given.

The server side output then provides better insights on the burst rate 
and the duty cycle (DC), as well as the end to end xfer time.

Burst
=====

root@raspberrypi:~# iperf -s -i 1 -e
------------------------------------------------------------
Server listening on TCP port 5001 with pid 2241
Read buffer size:  128 KByte (Dist bin width=16.0 KByte)
TCP congestion control default cubic
TCP window size:  128 KByte (default)
------------------------------------------------------------
[  1] local 192.168.1.31%eth0 port 5001 connected with 192.168.1.32 port 
55512 (burst-period=1.00s) (trip-times) (sock=4) (peer 2.1.10-dev) 
(icwnd/mss/irtt=14/1448/213) on 2023-11-14 10:01:05.936 (PST)
[ ID] Burst (start-end)  Transfer     Bandwidth       XferTime  (DC%)    
  Reads=Dist          NetPwr
[  1] 0.00-0.01 sec  1.00 MBytes   812 Mbits/sec  10.338 ms (1%)    
100=95:5:0:0:0:0:0:09812
[  1] 1.00-1.01 sec  1.00 MBytes   852 Mbits/sec   9.849 ms (0.98%)    
144=141:3:0:0:0:0:0:010810
[  1] 2.00-2.01 sec  1.00 MBytes   817 Mbits/sec  10.268 ms (1%)    
118=109:4:5:0:0:0:0:09946
[  1] 0.00-3.00 sec  3.00 MBytes  8.38 Mbits/sec  
10.152/9.849/10.338/0.264 ms  362=345:12:5:0:0:0:0:0

root@raspberrypi:~# iperf -c 192.168.1.31 --burst-size 1M --trip-times 
-i 1 -e --burst-period 1 -t 3
------------------------------------------------------------
Client connecting to 192.168.1.31, TCP port 5001 with pid 84171 (1/0 
flows/load)
Write buffer size: 131072 Byte  Burst size: 1048576 Byte
Bursting: 1.00 MByte every 1.00 second(s)
TCP congestion control using cubic
TOS set to 0x0 (Nagle on)
TCP window size: 85.0 KByte (default)
Event based writes (pending queue watermark at 16384 bytes)
------------------------------------------------------------
[  1] local 192.168.1.32%eth0 port 55512 connected with 192.168.1.31 
port 5001 (prefetch=16384) (trip-times) (sock=3) 
(icwnd/mss/irtt=14/1448/245) (ct=0.41 ms) on 2023-11-14 10:01:05.936 
(PST)
[ ID] Interval        Transfer    Bandwidth       Write/Err  Rtry     
Cwnd/RTT(var)        NetPwr
[  1] 0.00-1.00 sec  1.00 MBytes  8.39 Mbits/sec  8/0         0      
237K/1374(122) us  763
[  1] 1.00-2.00 sec  1.00 MBytes  8.39 Mbits/sec  8/0         0      
236K/1043(138) us  1005
[  1] 2.00-3.00 sec  1.00 MBytes  8.39 Mbits/sec  8/0         0      
179K/1490(196) us  704
[  1] 0.00-3.01 sec  3.00 MBytes  8.36 Mbits/sec  24/0         0      
179K/2591(2348) us  403

Isoch
=====

root@raspberrypi:~# iperf -s -i 1 -e --histograms
------------------------------------------------------------
Server listening on TCP port 5001 with pid 2253
Read buffer size:  128 KByte (Dist bin width=16.0 KByte)
TCP congestion control default cubic
Enabled receive histograms bin-width=0.100 ms, bins=100000 (clients 
should use --trip-times)
TCP window size:  128 KByte (default)
------------------------------------------------------------
[  1] local 192.168.1.31%eth0 port 5001 connected with 192.168.1.32 port 
39188 (isoch) (trip-times) (sock=4) (peer 2.1.10-dev) 
(icwnd/mss/irtt=14/1448/214) on 2023-11-14 10:06:54.178 (PST)
[ ID] Interval        Transfer    Bandwidth    Burst Latency 
avg/min/max/stdev (cnt/size) NetPwr  Reads=Dist
[  1] 0.00-1.00 sec  2.37 MBytes  19.9 Mbits/sec  
0.666/0.448/2.084/0.210 ms (60/41501) 3739  432=430:2:0:0:0:0:0:0
[  1] 0.00-1.00 sec F8-PDF: 
bin(w=100us):cnt(60)=5:2,6:21,7:20,8:12,9:3,10:1,21:1 
(5.00/95.00/99.7%=6/9/21,Outliers=1,obl/obu=0/0) (2.084 
ms/1699985214.182410)
[  1] 1.00-2.00 sec  2.47 MBytes  20.7 Mbits/sec  
0.657/0.458/0.944/0.106 ms (60/43104) 3939  457=455:2:0:0:0:0:0:0
[  1] 1.00-2.00 sec F8-PDF: 
bin(w=100us):cnt(60)=5:4,6:14,7:21,8:17,9:3,10:1 
(5.00/95.00/99.7%=5/9/10,Outliers=0,obl/obu=0/0) (0.944 
ms/1699985215.281164)
[  1] 2.00-3.00 sec  2.41 MBytes  20.2 Mbits/sec  
0.734/0.548/1.047/0.111 ms (60/42095) 3443  197=149:18:24:6:0:0:0:0
[  1] 2.00-3.00 sec F8-PDF: 
bin(w=100us):cnt(60)=6:5,7:17,8:23,9:10,10:3,11:2 
(5.00/95.00/99.7%=6/10/11,Outliers=0,obl/obu=0/0) (1.047 
ms/1699985216.881204)
[  1] 0.00-3.00 sec  7.29 MBytes  20.4 Mbits/sec  
0.686/0.448/2.084/0.153 ms (181/42227) 3712  1088=1035:22:25:6:0:0:0:0
[  1] 0.00-3.00 sec F8(f)-PDF: 
bin(w=100us):cnt(181)=5:6,6:40,7:58,8:53,9:16,10:5,11:2,21:1 
(5.00/95.00/99.7%=6/9/21,Outliers=1,obl/obu=0/0) (2.084 
ms/1699985214.182410)


root@raspberrypi:~# iperf -c 192.168.1.31 --burst-size 1M --trip-times 
-i 1 -e --isochronous=60:20m,4m -t 3
------------------------------------------------------------
Client connecting to 192.168.1.31, TCP port 5001 with pid 84182 (1/0 
flows/load)
Write buffer size: 131072 Byte  Burst size: 1048576 Byte
Isochronous: 60.00 frames/sec mean=20.0 Mbit/s, stddev=4.00 Mbit/s, 
Period/IPG=16.67/0.000 ms
TCP congestion control using cubic
TOS set to 0x0 (Nagle on)
TCP window size: 85.0 KByte (default)
Event based writes (pending queue watermark at 16384 bytes)
------------------------------------------------------------
[  1] local 192.168.1.32%eth0 port 39188 connected with 192.168.1.31 
port 5001 (prefetch=16384) (isoch) (trip-times) (sock=3) 
(icwnd/mss/irtt=14/1448/259) (ct=0.42 ms) on 2023-11-14 10:06:54.178 
(PST)
[ ID] Interval        Transfer     Bandwidth      Write/Err  Rtry     
Cwnd/RTT     isoch:tx/miss/slip  NetPwr
[  1] 0.00-1.00 sec  2.37 MBytes  19.9 Mbits/sec  60/0        0      
124K/525 us         61/0/04743
[  1] 1.00-2.00 sec  2.47 MBytes  20.7 Mbits/sec  60/0        0      
124K/479 us         60/0/05399
[  1] 2.00-3.00 sec  2.41 MBytes  20.2 Mbits/sec  60/0        0      
134K/578 us         60/0/04370
[  1] 0.00-3.04 sec  7.29 MBytes  20.1 Mbits/sec  181/0        0      
134K/596 us        182/0/04221
[  1] Isoch schedule errors (mean/min/max/stdev) = 
0.079/0.057/0.110/0.005 ms


Bob
> On Tue, Nov 14, 2023 at 12:25 PM Vint Cerf via Nnagain
> <nnagain@lists.bufferbloat.net> wrote:
>> 
>> if they had not been all together they would have been consuming tons 
>> of video capacity doing video conference calls....
> 
> There were plenty of remote attendees. One feed per room = ?
> 
> Videoconferencing capacity use is another statistic that is not really
> widely available. I do not know of any "modern" videoconferencing
> service that makes available bandwidth, loss, frame rate, frame size,
> "quality" statistics for any given session.Do they? Does IETF
> meetecho?  Is there a federal reporting requirement?
> 
>  I primarily use galene.org: that does all that for me (since I have
> source code and have had my fingers in videoconferencing apps for many
> years, and every time I get a few minutes try to improve "gcc"
> further. A decent frame rate is usually around 500k/sec on the up,
> 750k peak, and describing what happens on the down and how it scales
> by user beyond the scope of what I want to write today..
> 
> facetime is the videoconferencing "pig" at up to 4mb/sec.
> 
> ?
> 
> 
>> 
>> :-))
>> v
>> 
>> 
>> On Tue, Nov 14, 2023 at 10:46 AM Livingood, Jason via Nnagain 
>> <nnagain@lists.bufferbloat.net> wrote:
>>> 
>>> On the subject of how much bandwidth does one household need, here's 
>>> a fun stat for you.
>>> 
>>> 
>>> 
>>> At the IETF’s 118th meeting last week (Nov 4 – 10, 2023), there were 
>>> over 1,000 engineers in attendance. At peak there were 870 devices 
>>> connected to the WiFi network. Peak bandwidth usage:
>>> 
>>> Downstream peak ~750 Mbps
>>> Upstream ~250 Mbps
>>> 
>>> 
>>> 
>>> From my pre-meeting Twitter poll 
>>> (https://twitter.com/jlivingood/status/1720060429311901873):
>>> 
>>> _______________________________________________
>>> Nnagain mailing list
>>> Nnagain@lists.bufferbloat.net
>>> https://lists.bufferbloat.net/listinfo/nnagain
>> 
>> 
>> 
>> --
>> Please send any postal/overnight deliveries to:
>> Vint Cerf
>> Google, LLC
>> 1900 Reston Metro Plaza, 16th Floor
>> Reston, VA 20190
>> +1 (571) 213 1346
>> 
>> 
>> until further notice
>> 
>> 
>> 
>> _______________________________________________
>> Nnagain mailing list
>> Nnagain@lists.bufferbloat.net
>> https://lists.bufferbloat.net/listinfo/nnagain