AW: [e2e] Queue size of routers

[Fred L. Templin]

Alexandre,

Alexandre L. Grojsgold wrote:
> My personal assumptions:
> 
> a) the router software assigns, for each outgoing interface, a relatively
>    small buffer space, just enough to hold N packtes in queue ( N a number
>    between 10 and 100);
> 
> b) the buffer length is constant and independent of the traffic load. No
>    attempt is made by the router to increase the buffer length under heavy
>    load or when it senses (?) a long link with large RTT;
> 
> c) the maximum queue length probably independs of the interface
>    speed or other particular characteristic;
> 
> d) No attempt is made to avoid packet loss on congested links by putting
>    more buffer space for queueing on it.

Isn't the model you've articulated above exactly the model used for
queueing packets to be sent on link layer interfaces in the general
sense? I.e., network device drivers generally provide a small output
queue that is produced by the network (IP) layer and consumed by the
link layer interface. Then, as you say, packet loss occurs when links
are congested and/or the link layer interface cannot keep up with the
rate offered by the network layer. Am I understanding this correctly?

Thanks,

Fred Templin
ftemplin at iprg.nokia.com