[c-nsp] 100 meg throughput

[Jared Mauch]

On Wed, Dec 14, 2005 at 10:51:34PM -0500, Alex Rubenstein wrote:
> 
> int x/y
>  load 30
> 
>   30 second input rate 191000 bits/sec, 151 packets/sec
>   30 second output rate 6000 bits/sec, 1 packets/sec

	once again, this isn't real utilization.

	it's the result of this:

  new average = ((average - interval) * exp (-t/C)) + interval

(see http://www.cisco.com/en/US/products/sw/iosswrel/ps1818/products_tech_note09186a0080191323.shtml)

	I can type load 30 just as well as anyone else, but you don't
get to see the data you get on a Juniper.

	I can run a test stream through a device and watch it
the traffic there after a few seconds via sh int or monitor interface.

	It's not to say that the data isn't accurate for what it's meant
to represent, it just doesn't represent what I generally want to know.

	It's a lot easier to see those bursts when you're not
looking over a 30s time period in the CLI without doing aggressive
SNMP polling which drives up the cpu usage a lot more when looking at
all the interfaces vs the one you care about..

	- jared

> On Wed, 14 Dec 2005, Jeff Kell wrote:
> 
> >Jared Mauch wrote:
> >
> >>	I really wish Cisco would make better counters available on this
> >>(like you can see on a Juniper, the stats are not weighted over 300
> >>seconds).
> >>
> >They sort of did with CatOS... 'show traffic' would give you peaks (and
> >timestamps) on the backplane segments.  But I don't recall the interval
> >the peak refers to.
> >
> >Jeff
> >_______________________________________________
> >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/
> >
> 
> -- 
> Alex Rubenstein, AR97, K2AHR, alex at nac.net, latency, Al Reuben
> Net Access Corporation, 800-NET-ME-36, http://www.nac.net

-- 
Jared Mauch  | pgp key available via finger from jared at puck.nether.net
clue++;      | http://puck.nether.net/~jared/  My statements are only mine.