Re: [LARTC] HTB and theory

[Stef Coene <stef.coene@docum.org>]

On Monday 09 December 2002 22:45, Catalin Bucur wrote:
> -----BEGIN PGP SIGNED MESSAGE-----
> Hash: SHA1
>
> Stef Coene wrote:
> | Yes  : child ceil <= ceil of parent (I just added it to the faq page)
> |
> | Stef
>
> So far, so good. But what if I want to give my clients a little more
> bandwidth than (parent ceil) / (no. of clients)?
> Let's suppose that I have 10 clients, and guaranteed bandwidth is
> 5000Kbit. The parent ceil is 5000Kbit, childs' ceils are equal with
> 500Kbit every each. But I want to give - for example - 800Kbit for each
> one, so the parent ceil must be equal or greater than 8000Kbit. That is
> the problem, I don't think is a such good ideea to give the parent a
> ceil that is greater than guaranteed bandwidth; the result could be
> unpredictable.
The rule is :
 child ceil <= ceil of parent
So a child with ceil = 800 kbit and a parent with ceil = 5000 kbit is no 
problem.

Or are you speaking about rates :
  sum (rate child ) = rate of parent
You don't have to follow this rule, but it will make things more 
unpredictable.

Stef

-- 

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

_______________________________________________
LARTC mailing list / LARTC@mailman.ds9a.nl
http://mailman.ds9a.nl/mailman/listinfo/lartc HOWTO: http://lartc.org/