Reducing select() usage under load

David Nicol davidnicol at
Tue May 16 17:49:09 UTC 2006

On 5/12/06, Aaron Hopkins <lists at> wrote:

> available between perfect fairness and only answering requests from one
> socket when it is flooded.

Is a flood even realistic, given expected CPU speed and pipe widths?
It seems that the bottleneck would get hit on bandwidth before starvation
effects happen.  Could a flooding attack really fill the packet queue faster
than than the code can drain it?

And given this attack, processing a hundred packets on the attacked port
before getting to an unattacked port -- there's a tradeoff there to.

Mindlessly defending the status quo even as it shifts,

David Nicol

More information about the nsd-users mailing list