Hi Tony,
On 17/07/2026 19:20, Tony Li wrote:
Hi Peter,
##PP
We do not react to spikes. And we do incremental offload. All to avoid what
you described.
Then you are not solving the problem.
##PP2
we are, but maybe a different one that you intended to solve with
TTE. DCM is not trying to solve the microburst problem.
Long term congestion is better handled by bandwidth management, either
at the head ends in a distributed solution or by the controller in a
centralized solution. True microbursts are handled by output queues or
methods that are even faster than TTE or DCM. The problem at hand is
to deal with the mid-term congestion, in anticipating and mitigating
congestion until the larger mechanisms can react appropriately.
What problem do you think you’re solving?
not sure what is the exact categorization. DCM is trying to solve the
congestion that lasts at least several 10s of seconds and longer.
There are many networks that do not use controllers and do not use TE
either. That is where DCM helps. It is not meant to replace the
controller role, but it can help mitigate longer term congestion.
thanks,
Peter
Tony
_______________________________________________
Lsr mailing list -- [email protected]
To unsubscribe send an email to [email protected]
