Shady areas of TCP window autotuning?
John Schnizlein
schnizlein at isoc.org
Tue Mar 17 16:54:29 UTC 2009
Or use a transmission-layer protocol that optimizes delay end-to-end.
http://tools.ietf.org/html/draft-shalunov-ledbat-congestion-00
On 2009Mar17, at 12:47 PM, Joe Maimon wrote:
>
> Leo Bicknell wrote:
>
>> TCP needs drops to manage to the right speed.
>
> This is whats bad. TCP should be slightly more intelligent and start
> considering rtt jitter as its primary source of congestion
> information.
>
> Designing L2 network performance to optimize a l3 protocol is
> backwards.
>
More information about the NANOG
mailing list