flent
flent copied to clipboard
tohojo
Observations on Flent output for "Coffee Shop Bloat Tests"
Using Flent 1.9.9-git-2a08dd9:
I have been looking hard at Flent results for Dave's tests for coffee shops (my latest script.) I have a few fussy observations about the Flent output for the tests:
-
I like how the rrul and rrul_be tests summarize average and sum as the last items in their section. My eye is drawn to the discontinuity in the labels and sees the sum (after all is - that's the only number people care about :-) at the bottom. In the tcp_nup and tcp_ndown tests, it occurs to me to place the average & sum at the bottom of those lists as well.
-
I believe (but am guessing) that the rrul and rrul_be "TCP totals" row is the sum of the TCP download sum and the TCP upload sum. It occurs to me that statistic could be placed at the very bottom of the report instead of being "in the middle".
-
It would be helpful not to add a closing "." after the filename in the "Data file written to..." line. That makes it easier to select the file name for pasting into
flent --gui ...command. Mistakenly copying the "." doesn't seem to hurt, but I perseverate about having that "." in the file name.
As always, Flent is a great system for collecting and displaying reproducible data. Thanks.
bash-3.2$ sh coffee-shop-bloat-test.sh
Testing from Richs-MBP-10337.lan to flent-fremont.bufferbloat.net
Measuring baseline latency to 1.1.1.1...
1.1.1.1 : xmt/rcv/%loss = 5/5/0%, min/avg/max = 14.6/15.9/16.7
Started Flent 1.9.9-git-2a08dd9 using Python 3.7.3.
Starting tcp_ndown test. Expected run time: 70 seconds.
Data file written to ./tcp_ndown-2019-09-15T065532.373197.Richs-MBP-10337_lan.flent.gz.
Summary of tcp_ndown test run from 2019-09-15 10:55:32.373197
Title: 'Richs-MBP-10337.lan'
avg median # data pts
Ping (ms) ICMP : 103.47 102.00 ms 350
Ping (ms) ICMP 1.1.1.1 (extra) : 27.10 26.25 ms 350
TCP download avg : 1.48 N/A Mbits/s 353
TCP download sum : 5.93 N/A Mbits/s 353
TCP download::1 : 1.56 1.51 Mbits/s 353
TCP download::2 : 1.53 1.49 Mbits/s 353
TCP download::3 : 1.42 1.42 Mbits/s 353
TCP download::4 : 1.42 1.42 Mbits/s 353
Started Flent 1.9.9-git-2a08dd9 using Python 3.7.3.
Starting tcp_nup test. Expected run time: 70 seconds.
Data file written to ./tcp_nup-2019-09-15T065648.202997.Richs-MBP-10337_lan.flent.gz.
Summary of tcp_nup test run from 2019-09-15 10:56:48.202997
Title: 'Richs-MBP-10337.lan'
avg median # data pts
Ping (ms) ICMP : 103.12 101.00 ms 350
Ping (ms) ICMP 1.1.1.1 (extra) : 25.44 23.60 ms 350
TCP upload avg : 0.13 N/A Mbits/s 354
TCP upload sum : 0.52 N/A Mbits/s 354
TCP upload::1 : 0.14 0.16 Mbits/s 354
TCP upload::2 : 0.14 0.16 Mbits/s 354
TCP upload::3 : 0.14 0.15 Mbits/s 354
TCP upload::4 : 0.10 0.10 Mbits/s 354
Started Flent 1.9.9-git-2a08dd9 using Python 3.7.3.
Starting rrul test. Expected run time: 70 seconds.
Data file written to ./rrul-2019-09-15T065804.053659.Richs-MBP-10337_lan.flent.gz.
Summary of rrul test run from 2019-09-15 10:58:04.053659
Title: 'Richs-MBP-10337.lan'
avg median # data pts
Ping (ms) ICMP : 106.25 105.00 ms 350
Ping (ms) ICMP 1.1.1.1 (extra) : 31.23 30.55 ms 350
Ping (ms) UDP BE : 9.43 106.38 ms 352
Ping (ms) UDP BK : 9.40 106.72 ms 352
Ping (ms) UDP EF : 9.38 106.44 ms 352
Ping (ms) avg : 9.40 N/A ms 352
TCP download BE : 1.52 1.55 Mbits/s 352
TCP download BK : 1.52 1.52 Mbits/s 352
TCP download CS5 : 1.51 1.50 Mbits/s 352
TCP download EF : 1.47 1.53 Mbits/s 352
TCP download avg : 1.50 N/A Mbits/s 352
TCP download sum : 6.02 N/A Mbits/s 352
TCP totals : 6.45 N/A Mbits/s 352
TCP upload BE : 0.10 0.10 Mbits/s 352
TCP upload BK : 0.11 0.11 Mbits/s 352
TCP upload CS5 : 0.11 0.10 Mbits/s 352
TCP upload EF : 0.11 0.11 Mbits/s 352
TCP upload avg : 0.11 N/A Mbits/s 352
TCP upload sum : 0.43 N/A Mbits/s 352
Started Flent 1.9.9-git-2a08dd9 using Python 3.7.3.
Starting rrul_be test. Expected run time: 70 seconds.
Data file written to ./rrul_be-2019-09-15T065924.850091.Richs-MBP-10337_lan.flent.gz.
Summary of rrul_be test run from 2019-09-15 10:59:24.850091
Title: 'Richs-MBP-10337.lan'
avg median # data pts
Ping (ms) ICMP : 106.67 106.00 ms 350
Ping (ms) ICMP 1.1.1.1 (extra) : 30.96 30.30 ms 350
Ping (ms) UDP BE1 : 9.47 106.33 ms 352
Ping (ms) UDP BE2 : 9.40 107.30 ms 352
Ping (ms) UDP BE3 : 9.26 104.88 ms 352
Ping (ms) avg : 9.38 N/A ms 352
TCP download BE : 1.38 1.39 Mbits/s 352
TCP download BE2 : 1.61 1.67 Mbits/s 352
TCP download BE3 : 1.55 1.61 Mbits/s 352
TCP download BE4 : 1.50 1.56 Mbits/s 352
TCP download avg : 1.51 N/A Mbits/s 352
TCP download sum : 6.04 N/A Mbits/s 352
TCP totals : 6.48 N/A Mbits/s 352
TCP upload BE : 0.11 0.11 Mbits/s 352
TCP upload BE2 : 0.11 0.11 Mbits/s 352
TCP upload BE3 : 0.11 0.11 Mbits/s 352
TCP upload BE4 : 0.11 0.10 Mbits/s 352
TCP upload avg : 0.11 N/A Mbits/s 352
TCP upload sum : 0.44 N/A Mbits/s 352
bash-3.2$
Yes, but have you got out to any coffee shops yet and had a little fun?
So far, it's: Bufferbloat: 98% IPv6: 0%
Yesterday I ran a test in a coffee shop here in lisbon, with my guitar nearby, and ended up playing music all afternoon upstairs with 6 ladies from london, and the owner supplying me free beer to stick around. I never got around to talking about the wifi problem he had....
(that particular shop is blue_skipper_coffee_lisbon, and I also did tests at lpc, cemita's (where qos is on) in davenport, SFO, and my last airbnb "Rosa"s)
On Sun, Sep 15, 2019 at 12:52 PM Rich Brown [email protected] wrote:
Using Flent 1.9.9-git-2a08dd9:
I have been looking hard at Flent results for Dave's tests for coffee shops (my latest script.) https://github.com/richb-hanover/coffee-shop-bloat-test I have a few fussy observations about the Flent output for the tests:
I like how the rrul and rrul_be tests summarize average and sum as the last items in their section. My eye is drawn to the discontinuity in the labels and sees the sum (after all is - that's the only number people care about :-) at the bottom. In the tcp_nup and tcp_ndown tests, it occurs to me to place the average & sum at the bottom of those lists as well.
I believe (but am guessing) that the rrul and rrul_be "TCP totals" row is the sum of the TCP download sum and the TCP upload sum. It occurs to me that statistic could be placed at the very bottom of the report instead of being "in the middle".
It would be helpful not to add a closing "." after the filename in the "Data file written to..." line. That makes it easier to select the file name for pasting into flent --gui ... command. Mistakenly copying the "." doesn't seem to hurt, but I perseverate about having that "." in the file name.
As always, Flent is a great system for collecting and displaying reproducible data. Thanks.
bash-3.2$ sh coffee-shop-bloat-test.sh Testing from Richs-MBP-10337.lan to flent-fremont.bufferbloat.net Measuring baseline latency to 1.1.1.1... 1.1.1.1 : xmt/rcv/%loss = 5/5/0%, min/avg/max = 14.6/15.9/16.7
Started Flent 1.9.9-git-2a08dd9 using Python 3.7.3. Starting tcp_ndown test. Expected run time: 70 seconds. Data file written to ./tcp_ndown-2019-09-15T065532.373197.Richs-MBP-10337_lan.flent.gz.
Summary of tcp_ndown test run from 2019-09-15 10:55:32.373197 Title: 'Richs-MBP-10337.lan'
avg median # data ptsPing (ms) ICMP : 103.47 102.00 ms 350 Ping (ms) ICMP 1.1.1.1 (extra) : 27.10 26.25 ms 350 TCP download avg : 1.48 N/A Mbits/s 353 TCP download sum : 5.93 N/A Mbits/s 353 TCP download::1 : 1.56 1.51 Mbits/s 353 TCP download::2 : 1.53 1.49 Mbits/s 353 TCP download::3 : 1.42 1.42 Mbits/s 353 TCP download::4 : 1.42 1.42 Mbits/s 353 Started Flent 1.9.9-git-2a08dd9 using Python 3.7.3. Starting tcp_nup test. Expected run time: 70 seconds. Data file written to ./tcp_nup-2019-09-15T065648.202997.Richs-MBP-10337_lan.flent.gz.
Summary of tcp_nup test run from 2019-09-15 10:56:48.202997 Title: 'Richs-MBP-10337.lan'
avg median # data ptsPing (ms) ICMP : 103.12 101.00 ms 350 Ping (ms) ICMP 1.1.1.1 (extra) : 25.44 23.60 ms 350 TCP upload avg : 0.13 N/A Mbits/s 354 TCP upload sum : 0.52 N/A Mbits/s 354 TCP upload::1 : 0.14 0.16 Mbits/s 354 TCP upload::2 : 0.14 0.16 Mbits/s 354 TCP upload::3 : 0.14 0.15 Mbits/s 354 TCP upload::4 : 0.10 0.10 Mbits/s 354 Started Flent 1.9.9-git-2a08dd9 using Python 3.7.3. Starting rrul test. Expected run time: 70 seconds. Data file written to ./rrul-2019-09-15T065804.053659.Richs-MBP-10337_lan.flent.gz.
Summary of rrul test run from 2019-09-15 10:58:04.053659 Title: 'Richs-MBP-10337.lan'
avg median # data ptsPing (ms) ICMP : 106.25 105.00 ms 350 Ping (ms) ICMP 1.1.1.1 (extra) : 31.23 30.55 ms 350 Ping (ms) UDP BE : 9.43 106.38 ms 352 Ping (ms) UDP BK : 9.40 106.72 ms 352 Ping (ms) UDP EF : 9.38 106.44 ms 352 Ping (ms) avg : 9.40 N/A ms 352 TCP download BE : 1.52 1.55 Mbits/s 352 TCP download BK : 1.52 1.52 Mbits/s 352 TCP download CS5 : 1.51 1.50 Mbits/s 352 TCP download EF : 1.47 1.53 Mbits/s 352 TCP download avg : 1.50 N/A Mbits/s 352 TCP download sum : 6.02 N/A Mbits/s 352 TCP totals : 6.45 N/A Mbits/s 352 TCP upload BE : 0.10 0.10 Mbits/s 352 TCP upload BK : 0.11 0.11 Mbits/s 352 TCP upload CS5 : 0.11 0.10 Mbits/s 352 TCP upload EF : 0.11 0.11 Mbits/s 352 TCP upload avg : 0.11 N/A Mbits/s 352 TCP upload sum : 0.43 N/A Mbits/s 352 Started Flent 1.9.9-git-2a08dd9 using Python 3.7.3. Starting rrul_be test. Expected run time: 70 seconds. Data file written to ./rrul_be-2019-09-15T065924.850091.Richs-MBP-10337_lan.flent.gz.
Summary of rrul_be test run from 2019-09-15 10:59:24.850091 Title: 'Richs-MBP-10337.lan'
avg median # data ptsPing (ms) ICMP : 106.67 106.00 ms 350 Ping (ms) ICMP 1.1.1.1 (extra) : 30.96 30.30 ms 350 Ping (ms) UDP BE1 : 9.47 106.33 ms 352 Ping (ms) UDP BE2 : 9.40 107.30 ms 352 Ping (ms) UDP BE3 : 9.26 104.88 ms 352 Ping (ms) avg : 9.38 N/A ms 352 TCP download BE : 1.38 1.39 Mbits/s 352 TCP download BE2 : 1.61 1.67 Mbits/s 352 TCP download BE3 : 1.55 1.61 Mbits/s 352 TCP download BE4 : 1.50 1.56 Mbits/s 352 TCP download avg : 1.51 N/A Mbits/s 352 TCP download sum : 6.04 N/A Mbits/s 352 TCP totals : 6.48 N/A Mbits/s 352 TCP upload BE : 0.11 0.11 Mbits/s 352 TCP upload BE2 : 0.11 0.11 Mbits/s 352 TCP upload BE3 : 0.11 0.11 Mbits/s 352 TCP upload BE4 : 0.11 0.10 Mbits/s 352 TCP upload avg : 0.11 N/A Mbits/s 352 TCP upload sum : 0.44 N/A Mbits/s 352 bash-3.2$
— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub https://github.com/tohojo/flent/issues/183?email_source=notifications&email_token=AHVNJP43X7322ZTKYIUA2QDQJYOVLA5CNFSM4IW2J6RKYY3PNVWWK3TUL52HS4DFUVEXG43VMWVGG33NNVSW45C7NFSM4HLNWJRQ, or mute the thread https://github.com/notifications/unsubscribe-auth/AHVNJPY2LOOKWJHTPJJ4XFDQJYOVLANCNFSM4IW2J6RA .
Flent-users mailing list [email protected] http://flent.org/mailman/listinfo/flent-users_flent.org
--
Dave Täht CTO, TekLibre, LLC http://www.teklibre.com Tel: 1-831-205-9740
That should all be fixable, at least for new tests. The summary output just prints data series in the order they are specified in the data file, so harder to re-sort old files...
On 15 September 2019 13:51:48 CEST, Rich Brown [email protected] wrote:
Using Flent 1.9.9-git-2a08dd9:
I have been looking hard at Flent results for Dave's tests for coffee shops (my latest script.) I have a few fussy observations about the Flent output for the tests:
I like how the rrul and rrul_be tests summarize average and sum as the last items in their section. My eye is drawn to the discontinuity in the labels and sees the sum (after all is - that's the only number people care about :-) at the bottom. In the tcp_nup and tcp_ndown tests, it occurs to me to place the average & sum at the bottom of those lists as well.
I believe (but am guessing) that the rrul and rrul_be "TCP totals" row is the sum of the TCP download sum and the TCP upload sum. It occurs to me that statistic could be placed at the very bottom of the report instead of being "in the middle".
It would be helpful not to add a closing "." after the filename in the "Data file written to..." line. That makes it easier to select the file name for pasting into
flent --gui ...command. Mistakenly copying the "." doesn't seem to hurt, but I perseverate about having that "." in the file name.As always, Flent is a great system for collecting and displaying reproducible data. Thanks.
bash-3.2$ sh coffee-shop-bloat-test.sh Testing from Richs-MBP-10337.lan to flent-fremont.bufferbloat.net Measuring baseline latency to 1.1.1.1... 1.1.1.1 : xmt/rcv/%loss = 5/5/0%, min/avg/max = 14.6/15.9/16.7 Started Flent 1.9.9-git-2a08dd9 using Python 3.7.3. Starting tcp_ndown test. Expected run time: 70 seconds. Data file written to ./tcp_ndown-2019-09-15T065532.373197.Richs-MBP-10337_lan.flent.gz. Summary of tcp_ndown test run from 2019-09-15 10:55:32.373197 Title: 'Richs-MBP-10337.lan' avg median # data pts Ping (ms) ICMP : 103.47 102.00 ms 350 Ping (ms) ICMP 1.1.1.1 (extra) : 27.10 26.25 ms 350 TCP download avg : 1.48 N/A Mbits/s 353 TCP download sum : 5.93 N/A Mbits/s 353 TCP download::1 : 1.56 1.51 Mbits/s 353 TCP download::2 : 1.53 1.49 Mbits/s 353 TCP download::3 : 1.42 1.42 Mbits/s 353 TCP download::4 : 1.42 1.42 Mbits/s 353 Started Flent 1.9.9-git-2a08dd9 using Python 3.7.3. Starting tcp_nup test. Expected run time: 70 seconds. Data file written to ./tcp_nup-2019-09-15T065648.202997.Richs-MBP-10337_lan.flent.gz. Summary of tcp_nup test run from 2019-09-15 10:56:48.202997 Title: 'Richs-MBP-10337.lan' avg median # data pts Ping (ms) ICMP : 103.12 101.00 ms 350 Ping (ms) ICMP 1.1.1.1 (extra) : 25.44 23.60 ms 350 TCP upload avg : 0.13 N/A Mbits/s 354 TCP upload sum : 0.52 N/A Mbits/s 354 TCP upload::1 : 0.14 0.16 Mbits/s 354 TCP upload::2 : 0.14 0.16 Mbits/s 354 TCP upload::3 : 0.14 0.15 Mbits/s 354 TCP upload::4 : 0.10 0.10 Mbits/s 354 Started Flent 1.9.9-git-2a08dd9 using Python 3.7.3. Starting rrul test. Expected run time: 70 seconds. Data file written to ./rrul-2019-09-15T065804.053659.Richs-MBP-10337_lan.flent.gz. Summary of rrul test run from 2019-09-15 10:58:04.053659 Title: 'Richs-MBP-10337.lan' avg median # data pts Ping (ms) ICMP : 106.25 105.00 ms 350 Ping (ms) ICMP 1.1.1.1 (extra) : 31.23 30.55 ms 350 Ping (ms) UDP BE : 9.43 106.38 ms 352 Ping (ms) UDP BK : 9.40 106.72 ms 352 Ping (ms) UDP EF : 9.38 106.44 ms 352 Ping (ms) avg : 9.40 N/A ms 352 TCP download BE : 1.52 1.55 Mbits/s 352 TCP download BK : 1.52 1.52 Mbits/s 352 TCP download CS5 : 1.51 1.50 Mbits/s 352 TCP download EF : 1.47 1.53 Mbits/s 352 TCP download avg : 1.50 N/A Mbits/s 352 TCP download sum : 6.02 N/A Mbits/s 352 TCP totals : 6.45 N/A Mbits/s 352 TCP upload BE : 0.10 0.10 Mbits/s 352 TCP upload BK : 0.11 0.11 Mbits/s 352 TCP upload CS5 : 0.11 0.10 Mbits/s 352 TCP upload EF : 0.11 0.11 Mbits/s 352 TCP upload avg : 0.11 N/A Mbits/s 352 TCP upload sum : 0.43 N/A Mbits/s 352 Started Flent 1.9.9-git-2a08dd9 using Python 3.7.3. Starting rrul_be test. Expected run time: 70 seconds. Data file written to ./rrul_be-2019-09-15T065924.850091.Richs-MBP-10337_lan.flent.gz. Summary of rrul_be test run from 2019-09-15 10:59:24.850091 Title: 'Richs-MBP-10337.lan' avg median # data pts Ping (ms) ICMP : 106.67 106.00 ms 350 Ping (ms) ICMP 1.1.1.1 (extra) : 30.96 30.30 ms 350 Ping (ms) UDP BE1 : 9.47 106.33 ms 352 Ping (ms) UDP BE2 : 9.40 107.30 ms 352 Ping (ms) UDP BE3 : 9.26 104.88 ms 352 Ping (ms) avg : 9.38 N/A ms 352 TCP download BE : 1.38 1.39 Mbits/s 352 TCP download BE2 : 1.61 1.67 Mbits/s 352 TCP download BE3 : 1.55 1.61 Mbits/s 352 TCP download BE4 : 1.50 1.56 Mbits/s 352 TCP download avg : 1.51 N/A Mbits/s 352 TCP download sum : 6.04 N/A Mbits/s 352 TCP totals : 6.48 N/A Mbits/s 352 TCP upload BE : 0.11 0.11 Mbits/s 352 TCP upload BE2 : 0.11 0.11 Mbits/s 352 TCP upload BE3 : 0.11 0.11 Mbits/s 352 TCP upload BE4 : 0.11 0.10 Mbits/s 352 TCP upload avg : 0.11 N/A Mbits/s 352 TCP upload sum : 0.44 N/A Mbits/s 352 bash-3.2$
Ah bugger, seems I was wrong and that the series are actually just being sorted alphabetically. So guess I'll have to figure out a way to define a sorting order that makes more sense...
I'm all for gnarly hacks... Is it possible to change the names to "TCP upload::avg" and "TCP upload::sum"? Or change the individual tests to "TCP upload-N" and the "TCP upload-avg" and "TCP upload-sum"?
(Or would the name change invalidate previous-to-new comparisons?)
And yes, I've been to my local coffee shop, and the owner is receptive. But he has his own "IT guy" who takes care of the network, so I wanna be really sure I know what I'm talking about before I suggest any changes.
Update: You can also tell me to get a life :-) The change to the order of appearance of the results is far less important to me than substituting a nearby host for the -H host. Thanks
PS Thanks, too, for removing the "."
Rich Brown [email protected] writes:
I'm all for gnarly hacks... Is it possible to change the names to "TCP upload::avg" and "TCP upload::sum"? Or change the individual tests to "TCP upload-N" and the "TCP upload-avg" and "TCP upload-sum"?
Nah, that won't work. For one thing, it would break compatibility with old data files (the names are also the keys used when finding the data to plot).
Also, to fix the "totals" issue on RRUL we need to sort non-alphabetically anyway. So I think specifying an arbitrary sort order is really needed. I'll look into that...
-Toke
Rich said: ... You can also tell me to get a life :-)
Really - I can live without this. This is the kind of request that might possibly be cool (perhaps), but is hardly critical. Given that it could take a bunch of time to implement, I would counsel that you wait until you got more votes of support before implementing it. (Unless you really want to see it, in which case, go ahead!)
Rich Brown [email protected] writes:
Rich said: ... You can also tell me to get a life :-)
Really - I can live without this. This is the kind of request that might possibly be cool (perhaps), but is hardly critical. Given that it could take a bunch of time to implement, I would counsel that you wait until you got more votes of support before implementing it. (Unless you really want to see it, in which case, go ahead!)
Well, I didn't say anything about when I would look into it ;)
I do think it's worthwhile, though, and not too hard to do, so I guess I'll just leave this issue around and fix it when time permits...