nanog mailing list archives
Re: Shaping on a large scale
From: Colin Alston <karnaugh () karnaugh za net>
Date: Sat, 31 Jan 2009 10:53:59 +0200
On 2009/01/30 07:56 PM bert hubert wrote:
In general, the Linux packet shaping infrastructure is overly powerful, if very weakly documented - despite the LARTC efforts.
"Overly powerful" is a strong word. Sure it has countless poorly documented features, but then it fails at even the the most trivial task at shaping ingress without having to shape your other egress interface.
Essentially, you have to throw hardware at it.FreeBSD notably has less of these problems, but neither are all that viable on large scale throughputs.
Current thread:
- Re: Shaping on a large scale, (continued)
- Re: Shaping on a large scale Nuno Vieira - nfsi telecom (Jan 30)
- Re: Shaping on a large scale Arie Vayner (Jan 30)
- RE: Shaping on a large scale Scott Berkman (Jan 30)
- RE: Shaping on a large scale C. Jon Larsen (Jan 30)
- RE: Shaping on a large scale Chris Caputo (Jan 30)
- RE: Shaping on a large scale Bruce Grobler (Jan 30)
- RE: Shaping on a large scale C. Jon Larsen (Jan 30)
- Re: Shaping on a large scale Matthew Moyle-Croft (Jan 30)
- Re: Shaping on a large scale Adrian Chadd (Jan 30)
- Re: Shaping on a large scale Rubens Kuhl (Jan 31)
- Re: Shaping on a large scale bert hubert (Jan 30)
- Re: Shaping on a large scale Colin Alston (Jan 31)