Flent-users discussion archives
 help / color / mirror / Atom feed
From: Dave Taht <dave.taht@gmail.com>
To: "tohojo/flent"
	<reply+AHVNJP3FJVPSTXEWYLUEJQN4SPN6VEVBNHHCF7XB3M@reply.github.com>
Cc: tohojo/flent <flent@noreply.github.com>,
	Comment <comment@noreply.github.com>,
	flent-users <flent-users@flent.org>
Subject: Re: [Flent-users] [tohojo/flent] Customized test displaying error (#203)
Date: Fri, 3 Apr 2020 20:25:34 -0700	[thread overview]
Message-ID: <CAA93jw4aePsYyDCBnbyYf4F0CK6O_s9g86Q31fOtx12nHixK7w@mail.gmail.com> (raw)
In-Reply-To: <tohojo/flent/issues/203/608963421@github.com>


[-- Attachment #1.1: Type: text/plain, Size: 2607 bytes --]

sudo python3 setup.py install
or
make install

flent is actually measuring rate differences, it does not actually assign
bandwidth. So your result - where the first flow grabs more bandwidth than
the others, is an artifact of your queuing (probably fifo and tcp cubic).
The first flow wins. In the case where you have a delay based tcp, the last
flow usually wins. In the case of a good AQM, all flows converge to a good,
roughly equal rate in rapid order.

Summary statistics are not a very useful way of looking at tcp behaviors
over time.

for more details on what's going on in the tcp stack, on linux, the tcp_nup
test can be revealing. This will capture internal tcp stats on this test.

flent -H wherever -t whatever --te=upload_streams=4 --socket-stats
--step-size .04 tcp_nup

On Fri, Apr 3, 2020 at 8:12 PM olg33 <notifications@github.com> wrote:

> Thanks Tohojo,
>
> I just cloned the repository and ran Flent using the run-Flent script.
> It's awesome! I tested several scenarios and in all cases it worked as
> expected. I just Have a question, how is the bandwidth assigned to each
> marked data stream? for instance, in one test I got the following bandwidth
> for these TCP streams:
>
> TCP download::0 (11) : 85.08 86.78 Mbits/s
> TCP download::1 (15) : 52.73 53.28 Mbits/s
> TCP download::1 (10) : 57.78 58.45 Mbits/s
>
> Are these values 86.78Mb/s, 53.28Mb/s and 58.45 Mb/s assigned ramdomly?
> What's the criteria to distribute the bandwitdth this way? or, can it be
> assigned via parameter, like a percentage of the total bandwidth available?
>
> Finally, I'd like to properly install the cloned version of Flent via
> "make install". Normally there is an autogen.sh and/or a configure script
> we run before proceeding with the compilation and file creation -make
> install- but I can't find these files in the cloned file system. Could you
> tell me where to find these files or if they have a different name?
>
> Thank you so much for all your kind help.
>
> —
> You are receiving this because you commented.
> Reply to this email directly, view it on GitHub
> <https://github.com/tohojo/flent/issues/203#issuecomment-608963421>, or
> unsubscribe
> <https://github.com/notifications/unsubscribe-auth/AHVNJP4R7BPSPRGEQ7PNB23RK2QOVANCNFSM4LS2ZHEQ>
> .
> _______________________________________________
> Flent-users mailing list
> Flent-users@flent.org
> http://flent.org/mailman/listinfo/flent-users_flent.org
>


-- 
Make Music, Not War

Dave Täht
CTO, TekLibre, LLC
http://www.teklibre.com
Tel: 1-831-435-0729

[-- Attachment #1.2: Type: text/html, Size: 3684 bytes --]

[-- Attachment #2: Type: text/plain, Size: 151 bytes --]

_______________________________________________
Flent-users mailing list
Flent-users@flent.org
http://flent.org/mailman/listinfo/flent-users_flent.org

  parent reply	other threads:[~2020-04-04  3:26 UTC|newest]

Thread overview: 22+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
     [not found] <tohojo/flent/issues/203@github.com>
2020-03-24 17:42 ` Toke Høiland-Jørgensen
2020-03-26 14:54 ` Toke Høiland-Jørgensen
2020-03-26 17:59 ` Toke Høiland-Jørgensen
2020-03-31 14:15 ` Toke Høiland-Jørgensen
2020-04-01 10:09 ` Toke Høiland-Jørgensen
2020-04-01 10:33   ` Sebastian Moeller
2020-04-01 11:10 ` Toke Høiland-Jørgensen
2020-04-03  8:30 ` Toke Høiland-Jørgensen
     [not found] ` <tohojo/flent/issues/203/608963421@github.com>
2020-04-04  3:25   ` Dave Taht [this message]
2020-04-06  9:22 ` Toke Høiland-Jørgensen
2020-04-08 21:31 ` Toke Høiland-Jørgensen
2020-04-28  9:38 ` Toke Høiland-Jørgensen
     [not found] ` <tohojo/flent/issues/203/620745795@github.com>
2020-04-28 17:28   ` Dave Taht
     [not found] ` <tohojo/flent/issues/203/620754293@github.com>
2020-04-28 17:50   ` Dave Taht
2020-04-28 18:00 ` Toke Høiland-Jørgensen
2020-04-28 19:10 ` Toke Høiland-Jørgensen
     [not found] ` <tohojo/flent/issues/203/620824256@github.com>
2020-04-28 20:16   ` Dave Taht
2020-04-28 20:51 ` Toke Høiland-Jørgensen
2020-04-29 13:07 ` Toke Høiland-Jørgensen
2020-04-29 17:14 ` Toke Høiland-Jørgensen
2020-04-30 16:01 ` Toke Høiland-Jørgensen
2020-04-30 16:01 ` Toke Høiland-Jørgensen

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

  List information: https://lists.flent.org/postorius/lists/flent-users.flent.org/

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=CAA93jw4aePsYyDCBnbyYf4F0CK6O_s9g86Q31fOtx12nHixK7w@mail.gmail.com \
    --to=dave.taht@gmail.com \
    --cc=comment@noreply.github.com \
    --cc=flent-users@flent.org \
    --cc=flent@noreply.github.com \
    --cc=reply+AHVNJP3FJVPSTXEWYLUEJQN4SPN6VEVBNHHCF7XB3M@reply.github.com \
    --subject='Re: [Flent-users] [tohojo/flent] Customized test displaying error (#203)' \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox