[c-nsp] Sup2T - poor netflow performance

[Phil Mayers]

On 27/03/13 11:42, Jiri Prochazka wrote:
> Sam,
>
> NDE yielding is not something I would like to use at this stage, because
> I'm not able to get atleast to the same exporting capabilities as with
> the old Sup720.
>
> I tried to limit CPU usage for NDE to 20%, which had an effect of
> droping majority of flows.
>
> As soon as I limit usage to 70% (both Sup and linecards), no flows are
> dropped, but the box is still dying.

If it wasn't so irritating/tragic, this would be hilarious. People have 
repeatedly said "sup2T has operationally useful netflow" and yet it 
turns out to be *less* capable than sup720.

Hopefully this is crappy software or bugs, rather than architectural.

I would jump on Cisco, *hard*, about this. They've sang the netflow 
benefits of sup2T from the rooftops.