All of lore.kernel.org
 help / color / mirror / Atom feed
From: Andy Furniss <andy.furniss@dsl.pipex.com>
To: lartc@vger.kernel.org
Subject: Re: [LARTC] Delay before shaping kicks in
Date: Wed, 01 Mar 2006 14:56:21 +0000	[thread overview]
Message-ID: <4405B615.3070405@dsl.pipex.com> (raw)

Steinar Pedersen wrote:
> I am shaping traffic very sucessfully, but I still have one problem that I
> just can't figure out the solution to.
> When my backbone becomes saturated with traffic, I use prio to diffreciate
> between users, so that those who
> need it most, actually gets their service. BUT... when using prio, it
> actually takes a few seconds before the service they
> need is given to them.
> 
> An example:
> 
> One user downloads with 10mbit (full speed of the link)
> Another user with higher priority starts downloading, and for the first few
> seconds only get a fraction of what he really needs, but then suddenly, the
> speed increases, and the low-prio user looses his speed.
> 
> Is there a way to make this "kick in" start faster?
> 
> I checked the net/sched/sch_htb.c and found:
> #define HTB_EWMAC 2     /* rate average over HTB_EWMAC*HTB_HSIZE sec */
> 
> Could decreasing this to like 1 second fix the issue?

It's more likely to be because you are not allowing for overheads and 
have the rate on htb too high for the link so packets get queued elsewhere.

Andy.

_______________________________________________
LARTC mailing list
LARTC@mailman.ds9a.nl
http://mailman.ds9a.nl/cgi-bin/mailman/listinfo/lartc

             reply	other threads:[~2006-03-01 14:56 UTC|newest]

Thread overview: 2+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2006-03-01 14:56 Andy Furniss [this message]
  -- strict thread matches above, loose matches on Subject: below --
2006-02-26 20:39 [LARTC] Delay before shaping kicks in Steinar Pedersen

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=4405B615.3070405@dsl.pipex.com \
    --to=andy.furniss@dsl.pipex.com \
    --cc=lartc@vger.kernel.org \
    /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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.