netdev.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Rick Jones <rick.jones2@hp.com>
To: Volkan YAZICI <volkan.yazici@gmail.com>
Cc: netdev@vger.kernel.org
Subject: Re: How To Temporarily Suspend Network Traffic
Date: Tue, 01 Feb 2011 12:03:31 -0800	[thread overview]
Message-ID: <4D486713.70200@hp.com> (raw)
In-Reply-To: <87ipx335e1.fsf@alamut.ozu.edu.tr>

Volkan YAZICI wrote:
> On Tue, 01 Feb 2011 09:32:32 -0800, Rick Jones writes:
> 
>>Out of not quite idle curiousity, what are you trying to accomplish?
> 
> I'm trying to implement a coarse-grained soft-TDMA (Time Division
> Multiple Access) among devices in a Wi-Fi network. (Coarse-grained, that
> is, compared to a hardware implementation.) Assuming that device clocks
> are in sync via NTP, I will figure out the granularity I can achieve
> with soft-TDMA.

As in Fred gets to transmit from 0 to N, Ethel gets to transmit from N+1 to 2N 
and so on, based on absolute time?

Getting small number of microsecond synchronization between multiple systems via 
NTP (particularly if they are synchronized via a wireless network) may prove 
challenging.  At least that is my take as a member of the peanut gallery reading 
over the shoulders of the discussions that take place in comp.protocols.time.ntp

>>Instead of using tc to set a zero rate, you could perhaps try using tc
>>to set a delay? If it doesn't offer microseconds of delay, pehaps
>>setting a delay and then eliminating it after your own pause will do
>>what you want - depends of course on what it is you really want.
> 
> Thanks for the advice. I'll check this out and see what I can do.
> 
>>Your saying you wanted microsecond granularity suggests you don't want
>>to suspend traffic for very long?
> 
> I want to figure out the smallest delay that I can achieve in a periodic
> manner. (E.g., freeze the traffic for 200us, continue without
> interruption for 800us, and freeze the traffic again for another 200us,
> etc.) Pay attention that, an undeterministic delay somewhere in between
> 1000us is not something I prefer, I must be able to determine the point
> in time where delay will appear -- ofcourse, with some error margin.

Sounds, well, challenging :)  The determining the point in time part in 
particular.  One thing I've learned is that the cell towers in a cell network 
sync their time directly to GPS with some kit that is non-trivially expensive.

rick jones

> 
> 
> Regards.
> --
> To unsubscribe from this list: send the line "unsubscribe netdev" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html


      reply	other threads:[~2011-02-01 20:03 UTC|newest]

Thread overview: 6+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2011-02-01 14:28 How To Temporarily Suspend Network Traffic Volkan YAZICI
2011-02-01 17:32 ` Rick Jones
2011-02-01 17:49   ` Stephen Hemminger
2011-02-01 18:22     ` Volkan YAZICI
2011-02-01 18:12   ` Volkan YAZICI
2011-02-01 20:03     ` Rick Jones [this message]

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

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

  git send-email \
    --in-reply-to=4D486713.70200@hp.com \
    --to=rick.jones2@hp.com \
    --cc=netdev@vger.kernel.org \
    --cc=volkan.yazici@gmail.com \
    /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
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).