[j-nsp] RPM for performance monitoring?

[Richmond, Jeff]

I am using these today, on both M-series and J-series and they work relatively well. There are some caveats when dealing with Jitter, however, so I would suggest that you do test in your own lab to see if it will provide an acceptable solution. I had a very good JTAC engineer work with me to work through some of those issues and I think it ended up being a learning experience for both of us, but at the end of the day we finally ended up with a good workable Jitter solution.

Thanks,
-Jeff
________________________________________
From: juniper-nsp-bounces at puck.nether.net [juniper-nsp-bounces at puck.nether.net] On Behalf Of Nalkhande Tarique Abbas [ntarique at juniper.net]
Sent: Monday, September 14, 2009 9:09 AM
To: Stefan Fouant; Juniper List
Subject: Re: [j-nsp] RPM for performance monitoring?

I think the key differentiator here would be the RPM timestamps for
which you have two versions viz RE based timestamps or Hardware
timestamps.

The RE based version keeps timestamps in memory when packets are sent
and received.  This is not very accurate due to the delay added by PFE
to RE transit and waiting for CPU time. The desired accuracy for ICMP
Echo probes is on the order of that seen via the CLI 'ping' command

On the other side MS-PIC based time stamps provide the best performance
and accuracy since we have dedicated resources for time stamping.



Thanks & Regards,
Tarique A. Nalkhande

-----Original Message-----
From: juniper-nsp-bounces at puck.nether.net
[mailto:juniper-nsp-bounces at puck.nether.net] On Behalf Of Stefan Fouant
Sent: Monday, September 14, 2009 9:18 PM
To: Juniper List
Subject: [j-nsp] RPM for performance monitoring?

Folks,

I'm interested in gauging performance metrics between nodes (basically
looking for minimum, maximum, and average latency) and was wondering if
RPM
might be a suitable utility for measuring such performance.  Before I
take
an exhaustive look at this in the lab, I'd like to hear your
experiences.
Ideally, I'd like to push out an SNMP trap in the event the latency
exceeds
a certain threshold, but at a bare minimum I'll need to be able to look
at
this data in a historical context.

Any thoughts?

--
Stefan Fouant
_______________________________________________
juniper-nsp mailing list juniper-nsp at puck.nether.net
https://puck.nether.net/mailman/listinfo/juniper-nsp
_______________________________________________
juniper-nsp mailing list juniper-nsp at puck.nether.net
https://puck.nether.net/mailman/listinfo/juniper-nsp