Network SLA

isabel dias isabeldias1 at
Thu Feb 19 17:27:33 UTC 2009

Maybe the best way of addressing this is knowing exactly what we need to  measure- if IP traffic, services or processes. If the timescale of a process (ie: MTTR's)and/or procedure or just data and/or voice traffic from point A to B. Or just scoping the measurments as being the performance of the core network, or only related to usage based service. And that takes us to the TMN model and to the bottom-up approach starting w/ the FCAPs.

you have fereware, shareware and licenced tools or most likely specific vendor-related tools and only linked to one vendor or one type of equipment.  I am sure you've heard of RRD/MRTG, just like a few others that normally sit on the botton tier and have an upstream chain correlating the events. Most times the options are about suitablity and what the software version is prepared to report on so they are seen as more "suitable" to customers. 

--- On Thu, 2/19/09, Andreas, Rich <Rich_Andreas at> wrote:

> From: Andreas, Rich <Rich_Andreas at>
> Subject: RE: Network SLA
> To: "Saqib Ilyas" <msaqib at>, nanog at
> Date: Thursday, February 19, 2009, 5:59 PM
> Availability cannot be calculated in advance.  It typically
> is based on
> historical component failure information.  Sound design
> ensures
> redundancy and eliminates single point of failure.
> As for the rest, CIR, Latency, Jitter, Loss ..... this can
> be tested
> prior to customer handover with any number of tools and
> protocols
> including IEEE 802.11ag/ah, ITU-T 1731,  IETF RFC2544. 
> Hand-helds are
> typically not cost effective.  
> Rich Andreas
> Comcast Network Engineering
> -----Original Message-----
> From: Saqib Ilyas [mailto:msaqib at] 
> Sent: Thursday, February 19, 2009 10:50 AM
> To: nanog at
> Subject: Network SLA
> Greetings
> I am curious to know about any tools/techniques that a
> service provider
> uses
> to assess an SLA before signing it. That is to say, how
> does an
> administrator know if he/she can meet what he is promising.
> Is it based
> on
> experience? Are there commonly used tools for this?
> Thanks and best regards
> -- 
> Muhammad Saqib Ilyas
> PhD Student, Computer Science and Engineering
> Lahore University of Management Sciences


More information about the NANOG mailing list