At 16:27 -0500 12/14/97, hannan@uu.net wrote:
> The general rule-of-thumb metric I have used
> is that for every 60 packets of traffic, one
> ~1500 byte packet of netflow data will be generated.
>
> I won't attest to its accuracy, but empirically
> it's pretty close.
>
> -alan
presumably this relates to the duration of flows, so when new http's are
deployed that open and close fewer tcp's, the amount of export data will
decrease. It certainly will be interesting to monitor this.
dave
>Thus spake Jonathan Heiliger (loco@isi.net)
> on or about Sun, Dec 14, 1997 at 12:50:09PM -0800:
>> On Sun, 14 Dec 1997, Subodh Nijsure wrote:
>>
>> |} Does anyone know pps & bps rate for the *Cisco netflow export* data?
>> |} i.e. has anyone tracked utilization of an interface without netflow
>> |} export and with netflow export or anyone who has quantified netflow
>> |} export traffic as percentage of total traffic.
>>
>> We saw ~1-1.8M of hard-hitting NetFlow Export UDP traffic w/ ~60-80M of
>> input traffic on a router. Sorry, we didn't track bps/pps rates.
>>
>>
>> -jh-
This archive was generated by hypermail 2b29 : Sun Aug 04 2002 - 04:13:14 EDT