[Iccrg] Best congestion avoidance algorithm

[Saverio Mascolo]

On Wed, Jul 9, 2008 at 9:41 PM, Lachlan Andrew <lachlan.andrew at gmail.com> wrote:

  Greetings Adam,

  2008/7/9 adam maxiaodong <adam.maxiaodong at gmail.com>:

  >
  > But why fairness and friendliness are equally important?
  >
  > I though all those congestion algorithms are already intra-protocol fair
  > (otherwise, it will be a design flaw and won't get implemented)


  Even Reno isn't intra-protocol fair (unless RTTs are equal).
  Algorithms get implemented in Linux because someone takes the effort
  to implement them, not because they have been shown to be good.

  Lars is absolutely right that assessing based only on throughput is
  missing the point of congestion control.

what about measuring queuing delays? it would be another important metric!

best
saverio
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://oakham.cs.ucl.ac.uk/pipermail/iccrg/attachments/20080710/62f5d562/attachment.html