The previous answer I got from Dave O'leary and Greg Ketell [Juniper]
was that the hard limit is in place to minimize contention over the
aforementioned bus and writing to _disk_. Keep in mind that this bus is
used for other purposes and not just sampling as well.
-- steve
> -----Original Message-----
> From: Jesper Skriver [mailto:jesper@skriver.dk]
> Sent: Tuesday, December 11, 2001 1:28 PM
> To: Stephen Gill
> Cc: 'Krishna Doddapaneni'; juniper-nsp@puck.nether.net
> Subject: Re: [j-nsp] Traffic Sampling in JunOS5.1
>
> On Tue, Dec 11, 2001 at 01:16:44PM -0600, Stephen Gill wrote:
> > Limit is per box due to the RE - PFE bus (100MB) over which the
sampled
> > jcell must pass.
>
> Just doing a bit of math
>
> 64 bytes * 7000 pkts/s * 8 bit/byte < 3.6 Mbps
>
> So it's hardly a bandwidth problem, even though I've heard the
> same from Juniper employees, it's more likely a CPU problem on
> the routing engine ...
>
> /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.
_________________________________________________________
Do You Yahoo!?
Get your free @yahoo.com address at http://mail.yahoo.com
This archive was generated by hypermail 2b29 : Mon Aug 05 2002 - 10:42:38 EDT