[c-nsp] Nexus 5596 architecture

[John Gill]

Hi Jiri,
These total numbers are not a problem, all ports are equal and all 
traffic goes to the fabric on every port.  You will only see drops in 
this scenario if you have bursts of traffic going from many to one port 
for a period of time larger than the buffers will allow.  Remember, the 
buffer numbers are ingress, so if you have 2 ports sending to 1, you 
have 2x that buffer.  If you have 3 to 1, there are 3x buffers.

At those rates and if you know the burst, you can calculate how long 
your congestion can last.

Regards,
John Gill
cisco
919.392.2309

On 2/9/12 4:52 AM, Jiri Prochazka wrote:
> John,
>
> we are considering these nexus switches as a core for a small (for now)
> exchange point, so there will definitely be multiple ports talking to
> one and vice versa. Let's say the switch would be utilized up to 90% (45
> ports in case of 5548, 90 in case of 5596), half of the active ports
> would handle around 8 Gbps. The rest would be utilized up to 3 Gbps.
> Whole amount of a traffic would be standard Internet flows.
>
> This gives me real utilization of 240 Gbps or 500 Gbps in case of 5596.
> These are of course extreme values which will not be reached, but I need
> to know limits of this platform regarding to this use-case.
>
>
> Thanks,
>
>
> Jiri Prochazka
>