All of lore.kernel.org
 help / color / mirror / Atom feed
* [LARTC] TBF to limit bandwidth, Should I change #define HZ?
@ 2002-08-30 20:52 Fernando Lemos de Mello
  2002-08-31 11:28 ` bert hubert
  0 siblings, 1 reply; 2+ messages in thread
From: Fernando Lemos de Mello @ 2002-08-30 20:52 UTC (permalink / raw)
  To: lartc

Dear friends,

thanks for the last advices about TBF.

After reading the tbf.pdf document I got a better insight of the problems I
had trying to limit the bandwith on my testbed with a high resolution.

The question I have now is:
Would I get an immediatly better resolution result if I change the #define
HZ to 1000 for my Linux box? I have a Pentium Celeron 700 MHz, and I would
like to see the tbf working like:

send a 1000 bytes packet -> wait 10ms -> send a 1000 bytes packet -> wait
10ms ....

So if I change the file /usr/src/linux/include/asm-i386/param.h: #define HZ
100 to #define HZ 1000 and simply recompile the kernel, would tc work with
better resolution?

Thanks again,

Fernando.


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

^ permalink raw reply	[flat|nested] 2+ messages in thread

* Re: [LARTC] TBF to limit bandwidth, Should I change #define HZ?
  2002-08-30 20:52 [LARTC] TBF to limit bandwidth, Should I change #define HZ? Fernando Lemos de Mello
@ 2002-08-31 11:28 ` bert hubert
  0 siblings, 0 replies; 2+ messages in thread
From: bert hubert @ 2002-08-31 11:28 UTC (permalink / raw)
  To: lartc

On Fri, Aug 30, 2002 at 06:00:37PM -0300, Fernando Lemos de Mello wrote:
> Dear friends,
> 
> thanks for the last advices about TBF.
> 
> After reading the tbf.pdf document I got a better insight of the problems I
> had trying to limit the bandwith on my testbed with a high resolution.
> 
> The question I have now is:
> Would I get an immediatly better resolution result if I change the #define
> HZ to 1000 for my Linux box? I have a Pentium Celeron 700 MHz, and I would
> like to see the tbf working like:

You don't need to change HZ, for most purposes, it suffices to just raise
the burst size, which means 'send some packets' 'wait 10 ms' 'send some more
packets'. This will even out just fine.

> So if I change the file /usr/src/linux/include/asm-i386/param.h: #define HZ
> 100 to #define HZ 1000 and simply recompile the kernel, would tc work with
> better resolution?

Yes.

Kind regards,

bert hubert

-- 
http://www.PowerDNS.com          Versatile DNS Software & Services
http://www.tk                              the dot in .tk
http://lartc.org           Linux Advanced Routing & Traffic Control HOWTO
_______________________________________________
LARTC mailing list / LARTC@mailman.ds9a.nl
http://mailman.ds9a.nl/mailman/listinfo/lartc HOWTO: http://lartc.org/

^ permalink raw reply	[flat|nested] 2+ messages in thread

end of thread, other threads:[~2002-08-31 11:28 UTC | newest]

Thread overview: 2+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2002-08-30 20:52 [LARTC] TBF to limit bandwidth, Should I change #define HZ? Fernando Lemos de Mello
2002-08-31 11:28 ` bert hubert

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.