nanog mailing list archives
Re: Shady areas of TCP window autotuning?
From: Leo Bicknell <bicknell () ufp org>
Date: Wed, 18 Mar 2009 08:25:35 -0500
In a message written on Wed, Mar 18, 2009 at 09:04:42AM +0100, Marian ??urkovi?? wrote:
It's fine to have smaller buffers in the high-speed core, but at the edge you still need to buffer for full RTT if you want to fully utilize the link with TCP Reno. Thus my conclusion holds - if we reduce buffers at the bottleneck point to 50 msec, flows with RTT>50 msec would suffer from reduced throughput.
Ah, I understand your point now. There is a balance to be had at the edge; the tuning to support a single TCP stream at full bandwidth and the tuning to reduce latency and jitter are on some level incompatable; so one must strike a balance between the two. Of course...
Anyway we probably have no other chance in situations when the only available queueing is FIFO. And if this gets implemented on larger scale, it could even have a positive side-effect - it might finally motivate OS maintainers to seriously consider deploying some delay-sensitive variant of TCP since Reno will no longer give them the best results.
Many of the problems can be mitigated with well known queueing strategies. WRED, priority queues, and other options have been around long enough I refuse to believe they add significant cost. Rather, I think the problem is of awareness, far too few network engineers seem to really understand what effect the queuing options have on traffic. -- Leo Bicknell - bicknell () ufp org - CCIE 3440 PGP keys at http://www.ufp.org/~bicknell/
Attachment:
_bin
Description:
Current thread:
- Shady areas of TCP window autotuning? Marian Ďurkovič (Mar 16)
- Re: Shady areas of TCP window autotuning? David Andersen (Mar 16)
- Re: Shady areas of TCP window autotuning? Leo Bicknell (Mar 16)
- Re: Shady areas of TCP window autotuning? Wayne E. Bouchard (Mar 16)
- Re: Shady areas of TCP window autotuning? Lars Eggert (Mar 16)
- RE: Shady areas of TCP window autotuning? Frank Bulk - iName.com (Mar 16)
- Re: Shady areas of TCP window autotuning? Brett Frankenberger (Mar 17)
- Re: Shady areas of TCP window autotuning? Mikael Abrahamsson (Mar 17)
- Re: Shady areas of TCP window autotuning? Leo Bicknell (Mar 17)
- Re: Shady areas of TCP window autotuning? Marian Ďurkovič (Mar 18)
- Re: Shady areas of TCP window autotuning? Leo Bicknell (Mar 18)
- Re: Shady areas of TCP window autotuning? Marian Ďurkovič (Mar 17)
- Re: Shady areas of TCP window autotuning? Joe Maimon (Mar 17)
- Re: Shady areas of TCP window autotuning? John Schnizlein (Mar 17)
- Re: Shady areas of TCP window autotuning? Tony Finch (Mar 17)
- Re: Shady areas of TCP window autotuning? Lars Eggert (Mar 17)