One point of possible clarification,
all of the below is accurate in regards to our Netflow sampling. But the
new 5.1 feature is for "packet replication" out a PFE interface which sends
the entire packet and does so via the asics so doesn't have much of a
performance limit.
GK
At 10:09 AM 12/11/2001, Jesper Skriver wrote:
>On Tue, Dec 11, 2001 at 09:46:27AM -0800, Krishna Doddapaneni wrote:
> > Hi,
> > I was a little confused reading the Traffic Sampling documentation.Can
> > anyone familiar with JunOS clear my doubts.
> > 1)When you sample Traffic does the PFE send the whole packet or the
> > packet header only to CPU?
>
>I believe it's the first jcell, that is the first 64 bytes.
>
> > 2)And does the CPU exist on the Linecard or RE?
>
>The routing engine.
>
> > 3)The 7000pps limit that i saw on the mailing list ..is it per Box or
> > per Linecard?
>
>pr. box.
>
>/Jesper
>
>--
>Jesper Skriver, jesper(at)skriver(dot)dk - CCIE #5456
>Work: Network manager @ AS3292 (Tele Danmark DataNetworks)
>Private: FreeBSD committer @ AS2109 (A much smaller network ;-)
>
>One Unix to rule them all, One Resolver to find them,
>One IP to bring them all and in the zone to bind them.
This archive was generated by hypermail 2b29 : Mon Aug 05 2002 - 10:42:38 EDT