[LARTC] HTB quantum for small rates

Linux Advanced Routing and Traffic Control

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

 



On Sunday 22 December 2002 03:37, Hannes Ebner wrote:
> hi devik,
>
> i have found the following you have written some time ago:
> > It is important to know that for precise operation quantums need to
> > be as small as possible and larger than MTU. Normaly you don't need
> > to specify quantums manualy as HTB chooses precomputed values. It
> > computes classe's quantum (when you add or change it) as its rate
> > divided by r2q global parameter. Its default value is 10 and because
> > typical MTU is 1500 the default is good for rates from 15 kBps (120
> > kbit). For smaller minimal rates specify r2q 1 when creating qdisc -
> > it is good from 12 kbit which should be enough. If you will need you
> > can specify quantum manualy when adding or changing the class.
>
> what would be the best solution for rates smaller than 120kbit? i don't
> know how the higher rates are affected by scaling r2q down to 1, but
> would this be a solution for rates from 12kbit?
>
> would it be better to calculate the quantum for each rate or to set a
> small quantum for all rates?
The best you can do is following this rules and choose a good r2q :
  - 1500 < rate_small / r2q
  - rate_bigg / r2q < 60.000

with
  - rate_small =3D smallest rate
  - rate_bigg =3D biggest rate

Quantum is only used if you have 2 or more class fighting for remaining=20
bandwidth from the same parent.  And with remaining, I mean that each cla=
ss=20
already received it's rate and that the parent has some bandwidth left.

Stef

--=20

stef.coene@docum.org
 "Using Linux as bandwidth manager"
     http://www.docum.org/
     #lartc @ irc.oftc.net



[Index of Archives]     [LARTC Home Page]     [Netfilter]     [Netfilter Development]     [Network Development]     [Bugtraq]     [GCC Help]     [Yosemite News]     [Linux Kernel]     [Fedora Users]
  Powered by Linux