[c-nsp] Sup720 & netflow

[Phil Rosenthal]

Hi,

We have several 6500's with dfc3bxl on all blades.
We've found that with the netflow mask that drops all L7 data (just  
sends source/dest ip), we can handle approx 4-5 Gbit/sec per dfc3bxl  
of netflow data without drops.

It's far from perfect, but has been working mostly for us.

Our traffic is 99% HTTP.

--Phil

On Dec 5, 2006, at 4:04 PM, Tim Stevenson wrote:

> The rate is not really the concern, sup720 can collect NF stats at
> line rate. The question is the number of unique flows & the
> configured flow mask. Do you care about application layer data
> (protocol & L4 ports) or just source/dest or dest IP is enough?
>
> I think if you search the cisco-nsp archives you'll find several
> discussions of NF scalability on 6500/7600, including the balancing
> act between CPU/NF overflow/"real-time" stats.
>
> Tim
>
> At 02:45 PM 12/5/2006 -0500, Robert J. Adams noted:
>> Hello everyone,
>>
>> Is anyone running netflow (on sup720) w/ a serious amount of traffic?
>> We're pushing around 35-40G of traffic through a Sup720 and would  
>> like
>> to get an idea on which ASN's it's heading to. I've heard that  
>> anything
>>> 1G can be problematic?
>>
>> Thanks,
>> Robert Adams
>> _______________________________________________
>> cisco-nsp mailing list  cisco-nsp at puck.nether.net
>> https://puck.nether.net/mailman/listinfo/cisco-nsp
>> archive at http://puck.nether.net/pipermail/cisco-nsp/
>
>
>
> Tim Stevenson, tstevens at cisco.com
> Routing & Switching CCIE #5561
> Technical Marketing Engineer, Catalyst 6500
> Cisco Systems, http://www.cisco.com
> IP Phone: 408-526-6759
> ********************************************************
> The contents of this message may be *Cisco Confidential*
> and are intended for the specified recipients only.
> _______________________________________________
> cisco-nsp mailing list  cisco-nsp at puck.nether.net
> https://puck.nether.net/mailman/listinfo/cisco-nsp
> archive at http://puck.nether.net/pipermail/cisco-nsp/