Skip to contentSkip to conversation

cbq vs htb?

LARTC Griem, Hans T 3 participants 6 messages May 15, 2003 – May 22, 2003

lore
Share & tools
lore
#0 lore raw

Hello,

Does anyone know when one should use cbq versus (simpler more accurate) htb? Specifically does cbq have added functionality that may be of interest to certain applications?

from htb home> Both CBQ and HTB help you to control the use of the
outbound bandwidth on a given link. Both allow you to use one physical link to simulate several slower links and to send different kinds of traffic on different simulated links. In both cases, you have to specify how to divide the physical link into simulated links and how to decide which simulated link to use for a given
packet to be sent. http://luxik.cdi.cz/~devik/qos/htb/manual/userg.htm

Thanks for any insight,

Torsten
_______________________________________________
LARTC mailing list / [email protected]
http://mailman.ds9a.nl/mailman/listinfo/lartc HOWTO: http://lartc.org/

#1 lore raw
Quoted 14 lines · Hans T

Hello,

Does anyone know when one should use cbq versus (simpler more accurate) htb? Specifically does cbq have added functionality that may be of interest to certain applications?

from htb home> Both CBQ and HTB help you to control the use of the outbound bandwidth on a given link. Both allow you to use one physical link to simulate several slower links and to send different kinds of traffic on different simulated links. In both cases, you have to specify how to divide the physical link into simulated links and how to decide which simulated link to use for a given packet to be sent.
http://luxik.cdi.cz/~devik/qos/htb/manual/userg.htm

Quoted 1 line

Thanks for any insight,

- Cbq uses the physical link situation to caculate the rate. Example : if you want to send 1mbit on a 10mbit you need an idle time of 90% on the link. This can be a problem if you want to shape on ppp connection that can have a different link bandwidth.
- Htb is better documented (at least I have a better understanding of htb).
- Htb is active maintained.
- Cbq has some obscure options. And it's not always clear what they do.

Stef

Signature · 8 lines
[email protected]
 "Using Linux as bandwidth manager"
     http://www.docum.org/
     #lartc @ irc.oftc.net

_______________________________________________
LARTC mailing list / [email protected]
http://mailman.ds9a.nl/mailman/listinfo/lartc HOWTO: http://lartc.org/
#2 lore raw
In reply to Griem, Hans T

What I have told you, about the problem of ping delay with CBQ, is just with RF Conection..

Bye
Roberto.
----- Original Message -----
From: "Stef Coene" <[email protected]>
To: "Griem, Hans T" <[email protected]>
Cc: <[email protected]>
Sent: Thursday, May 15, 2003 6:26 PM
Subject: Re: [LARTC] cbq vs htb?

Quoted 4 lines · Hans T
Quoted 3 lines

Hello Cef or...,

Thanks for your input. Yes I am trying to figure out where/what/when

these

Quoted 1 line
Quoted 1 line

obscure CBQ options add value (ie., to conclude whether I should

eliminate

Quoted 1 line
Quoted 1 line

cbq from my "toolchest"). So I wonder since cbq uses the physical link

per

Quoted 3 lines
Quoted 2 lines

your response is it better suited to bandwidth control for (rf) applications with fluctuating link rates, etc?

In contrary. The bandwidth option in cbq should match the real physical

link

Quoted 2 lines

bandwidth. I don't know how this is done on rf networks. Comared this to htb. Htb uses a token bucket system to control the rate

of

Quoted 1 line

the data. This has nothing to do with the physical link. It just send

data

Quoted 8 lines

at a certain controlled rate.

I'm not a specialist in this. But I don't think I'm telling any lies :)

Stef

--

Quoted 10 lines

[email protected]
"Using Linux as bandwidth manager"
http://www.docum.org/
#lartc @ irc.oftc.net

_______________________________________________
LARTC mailing list / [email protected]
http://mailman.ds9a.nl/mailman/listinfo/lartc HOWTO: http://lartc.org/

_______________________________________________
LARTC mailing list / [email protected]
http://mailman.ds9a.nl/mailman/listinfo/lartc HOWTO: http://lartc.org/

#3 lore raw
In reply to Griem, Hans T

Hello Cef or...,

Thanks for your input. Yes I am trying to figure out where/what/when these obscure CBQ options add value (ie., to conclude whether I should eliminate cbq from my "toolchest"). So I wonder since cbq uses the physical link per your response is it better suited to bandwidth control for (rf) applications with fluctuating link rates, etc?

Torsten

-----Original Message-----
From: Stef Coene [mailto:[email protected]]
Sent: Thursday, May 15, 2003 10:00 AM
To: Griem, Hans T; [email protected]
Subject: Re: [LARTC] cbq vs htb?

Quoted 14 lines · Hans T

Hello,

Does anyone know when one should use cbq versus (simpler more accurate) htb? Specifically does cbq have added functionality that may be of interest to certain applications?

from htb home> Both CBQ and HTB help you to control the use of the outbound bandwidth on a given link. Both allow you to use one physical link to simulate several slower links and to send different kinds of traffic on different simulated links. In both cases, you have to specify how to divide the physical link into simulated links and how to decide which simulated link to use for a given packet to be sent.
http://luxik.cdi.cz/~devik/qos/htb/manual/userg.htm

Quoted 1 line

Thanks for any insight,

- Cbq uses the physical link situation to caculate the rate. Example : if you want to send 1mbit on a 10mbit you need an idle time of 90% on the link. This can be a problem if you want to shape on ppp connection that can have a different link bandwidth.
- Htb is better documented (at least I have a better understanding of htb).
- Htb is active maintained.
- Cbq has some obscure options. And it's not always clear what they do.

Stef

Signature · 8 lines
[email protected]
 "Using Linux as bandwidth manager"
     http://www.docum.org/
     #lartc @ irc.oftc.net

_______________________________________________
LARTC mailing list / [email protected]
http://mailman.ds9a.nl/mailman/listinfo/lartc HOWTO: http://lartc.org/
#4 lore raw
In reply to Griem, Hans T
Quoted 7 lines · Hans T

Hello Cef or...,

Thanks for your input. Yes I am trying to figure out where/what/when these obscure CBQ options add value (ie., to conclude whether I should eliminate cbq from my "toolchest"). So I wonder since cbq uses the physical link per your response is it better suited to bandwidth control for (rf) applications with fluctuating link rates, etc?

In contrary. The bandwidth option in cbq should match the real physical link bandwidth. I don't know how this is done on rf networks.
Comared this to htb. Htb uses a token bucket system to control the rate of the data. This has nothing to do with the physical link. It just send data at a certain controlled rate.

I'm not a specialist in this. But I don't think I'm telling any lies :)

Stef

Signature · 8 lines
[email protected]
 "Using Linux as bandwidth manager"
     http://www.docum.org/
     #lartc @ irc.oftc.net

_______________________________________________
LARTC mailing list / [email protected]
http://mailman.ds9a.nl/mailman/listinfo/lartc HOWTO: http://lartc.org/
#5 lore raw
In reply to Griem, Hans T
Quoted 2 lines · Intercom - Roberto Ravetti

What I have told you, about the problem of ping delay with CBQ, is just with RF Conection..

I don't know how a RF connections is physical seen by the kernel. Htb does not care about the physical link so maybe better suited to shape on a RF connection.

Stef

Signature · 8 lines
[email protected]
 "Using Linux as bandwidth manager"
     http://www.docum.org/
     #lartc @ irc.oftc.net

_______________________________________________
LARTC mailing list / [email protected]
http://mailman.ds9a.nl/mailman/listinfo/lartc HOWTO: http://lartc.org/