* [Make-wifi-fast] Netpower, iper2, BBR
@ 2018-08-04 17:55 Bob McMahon
2018-08-04 22:47 ` Jeremy Harris
0 siblings, 1 reply; 4+ messages in thread
From: Bob McMahon @ 2018-08-04 17:55 UTC (permalink / raw)
To: Make-Wifi-fast
[-- Attachment #1: Type: text/plain, Size: 6602 bytes --]
FYI, I've added a netpower metric to iperf 2.
Below are two runs over wired GigE comparing Google's BBR
<https://ai.google/research/pubs/pub45646> with cubic. Note, for the
network power the delay units are microseconds while the throughput are
seconds.
*Internet Access Runs (uplink)*
*A cubic run, netpower of 7.56*
[root@rjm-clubhouse-28 rjmcmahon]# echo "cubic" >
/proc/sys/net/ipv4/tcp_congestion_control
[root@rjm-clubhouse-28 rjmcmahon]# iperf -c mail.rjmcmahon.com -e -i 1
------------------------------------------------------------
Client connecting to mail.rjmcmahon.com, TCP port 5001 with pid 26458
Write buffer size: 128 KByte
TCP window size: 85.0 KByte (default)
------------------------------------------------------------
[ 3] local 192.168.100.10 port 40932 connected with 45.33.58.123 port 5001
[ ID] Interval Transfer Bandwidth Write/Err Rtry
Cwnd/RTT NetPwr
[ 3] 0.00-1.00 sec 2.62 MBytes 22.0 Mbits/sec 21/0 0
308K/200486 us 13.73
[ 3] 1.00-2.00 sec 2.00 MBytes 16.8 Mbits/sec 16/0 0
397K/258738 us 8.11
[ 3] 2.00-3.00 sec 1.48 MBytes 12.4 Mbits/sec 13/0 13
301K/198417 us 7.82
[ 3] 3.00-4.00 sec 1.49 MBytes 12.5 Mbits/sec 13/0 0
343K/229095 us 6.83
[ 3] 4.00-5.00 sec 1.43 MBytes 12.0 Mbits/sec 12/0 12
343K/243296 us 6.16
[ 3] 5.00-6.00 sec 1.49 MBytes 12.5 Mbits/sec 13/0 6
279K/187902 us 8.32
[ 3] 6.00-7.00 sec 1.43 MBytes 12.0 Mbits/sec 12/0 0
301K/204321 us 7.33
[ 3] 7.00-8.00 sec 1.37 MBytes 11.5 Mbits/sec 12/0 0
312K/213675 us 6.71
[ 3] 8.00-9.00 sec 1.43 MBytes 12.0 Mbits/sec 12/0 0
316K/219227 us 6.84
[ 3] 9.00-10.00 sec 1.43 MBytes 12.0 Mbits/sec 12/0 0
316K/216670 us 6.92
[ 3] 0.00-10.33 sec 16.2 MBytes 13.1 Mbits/sec 136/0 31
316K/217136 us 7.56
*A run with BBR, netpower 55.36*
[root@rjm-clubhouse-28 rjmcmahon]# echo "bbr" >
/proc/sys/net/ipv4/tcp_congestion_control
[root@rjm-clubhouse-28 rjmcmahon]# iperf -c mail.rjmcmahon.com -e -i 1
------------------------------------------------------------
Client connecting to mail.rjmcmahon.com, TCP port 5001 with pid 26499
Write buffer size: 128 KByte
TCP window size: 128 KByte (default)
------------------------------------------------------------
[ 3] local 192.168.100.10 port 40938 connected with 45.33.58.123 port 5001
[ ID] Interval Transfer Bandwidth Write/Err Rtry
Cwnd/RTT NetPwr
[ 3] 0.00-1.00 sec 2.88 MBytes 24.1 Mbits/sec 23/0 0
48K/28064 us 107.42
[ 3] 1.00-2.00 sec 1.50 MBytes 12.6 Mbits/sec 12/0 0
45K/29419 us 53.46
[ 3] 2.00-3.00 sec 1.00 MBytes 8.39 Mbits/sec 8/0 0
45K/29583 us 35.45
[ 3] 3.00-4.00 sec 1.50 MBytes 12.6 Mbits/sec 12/0 0
45K/27487 us 57.22
[ 3] 4.00-5.00 sec 1.38 MBytes 11.5 Mbits/sec 11/0 0
48K/31117 us 46.33
[ 3] 5.00-6.00 sec 1.50 MBytes 12.6 Mbits/sec 12/0 0
45K/19685 us 79.90
[ 3] 6.00-7.00 sec 1.50 MBytes 12.6 Mbits/sec 12/0 0
45K/29370 us 53.55
[ 3] 7.00-8.00 sec 1.50 MBytes 12.6 Mbits/sec 12/0 0
45K/27022 us 58.21
[ 3] 8.00-9.00 sec 1.50 MBytes 12.6 Mbits/sec 12/0 0
48K/31072 us 50.62
[ 3] 9.00-10.00 sec 1.00 MBytes 8.39 Mbits/sec 8/0 0
5K/31941 us 32.83
[ 3] 0.00-10.03 sec 15.2 MBytes 12.7 Mbits/sec 122/0 0
5K/28791 us 55.36
*Local Wired Wthernet Runs*
*A cubic run, netpower of 53532*
[root@rjm-clubhouse-28 iperf2-code]# uname -r
4.17.9-200.fc28.x86_64
[root@rjm-clubhouse-28 iperf2-code]# echo "cubic" >
/proc/sys/net/ipv4/tcp_congestion_control
[root@rjm-clubhouse-28 iperf2-code]# iperf -c 192.168.100.33 -e -i 1
------------------------------------------------------------
Client connecting to 192.168.100.33, TCP port 5001 with pid 17263
Write buffer size: 128 KByte
TCP window size: 85.0 KByte (default)
------------------------------------------------------------
[ 3] local 192.168.100.10 port 56218 connected with 192.168.100.33 port
5001
[ ID] Interval Transfer Bandwidth Write/Err Rtry
Cwnd/RTT NetPwr
[ 3] 0.00-1.00 sec 113 MBytes 947 Mbits/sec 903/0 0
373K/2231 us 53051.55
[ 3] 1.00-2.00 sec 112 MBytes 937 Mbits/sec 894/0 0
373K/2107 us 55613.84
[ 3] 2.00-3.00 sec 112 MBytes 937 Mbits/sec 894/0 0
373K/2205 us 53142.12
[ 3] 3.00-4.00 sec 112 MBytes 938 Mbits/sec 895/0 0
373K/2194 us 53468.30
[ 3] 4.00-5.00 sec 111 MBytes 932 Mbits/sec 889/0 0
373K/2181 us 53426.41
[ 3] 5.00-6.00 sec 112 MBytes 940 Mbits/sec 896/0 0
373K/2221 us 52877.31
[ 3] 6.00-7.00 sec 112 MBytes 940 Mbits/sec 896/0 0
373K/2198 us 53430.62
[ 3] 7.00-8.00 sec 111 MBytes 933 Mbits/sec 890/0 0
373K/2203 us 52952.37
[ 3] 8.00-9.00 sec 111 MBytes 934 Mbits/sec 891/0 0
373K/2169 us 53842.85
[ 3] 9.00-10.00 sec 112 MBytes 941 Mbits/sec 897/0 0
373K/2187 us 53759.30
[ 3] 0.00-10.01 sec 1.09 GBytes 937 Mbits/sec 8945/0 0
373K/2189 us 53532.01
*A run with BBR, netpower 72071*
[root@rjm-clubhouse-28 iperf2-code]# iperf -c 192.168.100.33 -e -i 1
------------------------------------------------------------
Client connecting to 192.168.100.33, TCP port 5001 with pid 18312
Write buffer size: 128 KByte
TCP window size: 128 KByte (default)
------------------------------------------------------------
[ 3] local 192.168.100.10 port 56238 connected with 192.168.100.33 port
5001
[ ID] Interval Transfer Bandwidth Write/Err Rtry
Cwnd/RTT NetPwr
[ 3] 0.00-1.00 sec 114 MBytes 956 Mbits/sec 912/0 0
274K/1644 us 72711.47
[ 3] 1.00-2.00 sec 111 MBytes 933 Mbits/sec 890/0 0
271K/1619 us 72053.17
[ 3] 2.00-3.00 sec 112 MBytes 937 Mbits/sec 894/0 0
274K/1720 us 68126.96
[ 3] 3.00-4.00 sec 112 MBytes 938 Mbits/sec 895/0 0
271K/1599 us 73364.25
[ 3] 4.00-5.00 sec 112 MBytes 938 Mbits/sec 895/0 0
274K/1603 us 73181.19
[ 3] 5.00-6.00 sec 111 MBytes 930 Mbits/sec 887/0 0
271K/1609 us 72256.60
[ 3] 6.00-7.00 sec 112 MBytes 940 Mbits/sec 896/0 0
274K/1579 us 74376.51
[ 3] 7.00-8.00 sec 112 MBytes 937 Mbits/sec 894/0 0
274K/1615 us 72556.27
[ 3] 8.00-9.00 sec 112 MBytes 935 Mbits/sec 892/0 0
274K/1646 us 71030.51
^C[ 3] 0.00-9.49 sec 1.04 GBytes 938 Mbits/sec 8489/0 0
271K/1627 us 72071.69
Bob
[-- Attachment #2: Type: text/html, Size: 21750 bytes --]
^ permalink raw reply [flat|nested] 4+ messages in thread
* Re: [Make-wifi-fast] Netpower, iper2, BBR
2018-08-04 17:55 [Make-wifi-fast] Netpower, iper2, BBR Bob McMahon
@ 2018-08-04 22:47 ` Jeremy Harris
2018-08-05 0:10 ` Bob McMahon
0 siblings, 1 reply; 4+ messages in thread
From: Jeremy Harris @ 2018-08-04 22:47 UTC (permalink / raw)
To: make-wifi-fast
On 08/04/2018 06:55 PM, Bob McMahon wrote:
> FYI, I've added a netpower metric to iperf 2.
>
> Below are two runs over wired GigE comparing Google's BBR
> <https://ai.google/research/pubs/pub45646> with cubic. Note, for the
> network power the delay units are microseconds while the throughput are
> seconds.
Why is it useful?
It appears to be calculated as bandwidth divided by latency, so the
units are bits / time^2. The word "power" in the naming feels
misleading.
--
Cheers,
Jeremy
^ permalink raw reply [flat|nested] 4+ messages in thread
* Re: [Make-wifi-fast] Netpower, iper2, BBR
2018-08-04 22:47 ` Jeremy Harris
@ 2018-08-05 0:10 ` Bob McMahon
2018-08-05 0:14 ` Bob McMahon
0 siblings, 1 reply; 4+ messages in thread
From: Bob McMahon @ 2018-08-05 0:10 UTC (permalink / raw)
To: Jeremy Harris; +Cc: Make-Wifi-fast
[-- Attachment #1.1: Type: text/plain, Size: 994 bytes --]
It's a metric to determine if network congestion is being optimized.
Here's just one of many papers that use. This paper says that flow
controlled, per network power, can't be optimized with local information
only.
Bob
On Sat, Aug 4, 2018 at 3:47 PM, Jeremy Harris <jgh@wizmail.org> wrote:
> On 08/04/2018 06:55 PM, Bob McMahon wrote:
> > FYI, I've added a netpower metric to iperf 2.
> >
> > Below are two runs over wired GigE comparing Google's BBR
> > <https://ai.google/research/pubs/pub45646> with cubic. Note, for the
> > network power the delay units are microseconds while the throughput are
> > seconds.
>
> Why is it useful?
>
> It appears to be calculated as bandwidth divided by latency, so the
> units are bits / time^2. The word "power" in the naming feels
> misleading.
> --
> Cheers,
> Jeremy
>
> _______________________________________________
> Make-wifi-fast mailing list
> Make-wifi-fast@lists.bufferbloat.net
> https://lists.bufferbloat.net/listinfo/make-wifi-fast
[-- Attachment #1.2: Type: text/html, Size: 1780 bytes --]
[-- Attachment #2: NoDistFlowControl.pdf --]
[-- Type: application/pdf, Size: 649623 bytes --]
^ permalink raw reply [flat|nested] 4+ messages in thread
* Re: [Make-wifi-fast] Netpower, iper2, BBR
2018-08-05 0:10 ` Bob McMahon
@ 2018-08-05 0:14 ` Bob McMahon
0 siblings, 0 replies; 4+ messages in thread
From: Bob McMahon @ 2018-08-05 0:14 UTC (permalink / raw)
To: Jeremy Harris; +Cc: Make-Wifi-fast
[-- Attachment #1.1: Type: text/plain, Size: 1168 bytes --]
Another interesting paper with a power metric.
Bob
On Sat, Aug 4, 2018 at 5:10 PM, Bob McMahon <bob.mcmahon@broadcom.com>
wrote:
> It's a metric to determine if network congestion is being optimized.
> Here's just one of many papers that use. This paper says that flow
> controlled, per network power, can't be optimized with local information
> only.
>
> Bob
>
> On Sat, Aug 4, 2018 at 3:47 PM, Jeremy Harris <jgh@wizmail.org> wrote:
>
>> On 08/04/2018 06:55 PM, Bob McMahon wrote:
>> > FYI, I've added a netpower metric to iperf 2.
>> >
>> > Below are two runs over wired GigE comparing Google's BBR
>> > <https://ai.google/research/pubs/pub45646> with cubic. Note, for the
>> > network power the delay units are microseconds while the throughput are
>> > seconds.
>>
>> Why is it useful?
>>
>> It appears to be calculated as bandwidth divided by latency, so the
>> units are bits / time^2. The word "power" in the naming feels
>> misleading.
>> --
>> Cheers,
>> Jeremy
>>
>> _______________________________________________
>> Make-wifi-fast mailing list
>> Make-wifi-fast@lists.bufferbloat.net
>> https://lists.bufferbloat.net/listinfo/make-wifi-fast
>
>
>
[-- Attachment #1.2: Type: text/html, Size: 2359 bytes --]
[-- Attachment #2: Internet Congestion Control Using the Power Metric-Keep the Pipe Just Full, But No Fuller July 2018.pdf --]
[-- Type: application/pdf, Size: 3845380 bytes --]
^ permalink raw reply [flat|nested] 4+ messages in thread
end of thread, other threads:[~2018-08-05 0:14 UTC | newest]
Thread overview: 4+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2018-08-04 17:55 [Make-wifi-fast] Netpower, iper2, BBR Bob McMahon
2018-08-04 22:47 ` Jeremy Harris
2018-08-05 0:10 ` Bob McMahon
2018-08-05 0:14 ` Bob McMahon
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox