ITIL Network Availability Measurement Calculation. Tricky?

General discussion on all aspects of the IT Infrastructure Library (ITIL)
Post Reply
User avatar
Posts: 10
Joined: Tue May 08, 2018 8:00 pm

Thu Jul 12, 2018 7:19 am

Yesterday we had thought provoking discussion on “How to Calculate Network Availability in below given scenario?” in our SIAM Expert Whatsapp group where we have 200+ experts..
Thought of sharing same with all of you…
Consider below Scenario and let me know is it fairly simple or tricky to calculate Network Availability SLA ?

1.NovelVista has several remote offices with their own Internet connection.

2.Each office connects back to the headquarters-HQ’s via a VPN tunnel.

3.Users at each remote office need to access a mix of internal applications
across the VPN tunnel and external services via the local Internet connection.

Here is my Question

1.What will be network availability value formula if the VPN tunnel is down, but not the Internet connection?

2.Will this availability value calculation formula change if the Internet connection goes down?

Solution -
We discussed this, brainstormed and got amazing answers… Here is summary of discussion
In normal situation we will calculate network availability based on the uptime of the Internet connection, because if the Internet goes down, it will affect both internal and external applications. But on other side in reality - a failure of the VPN tunnel would still impact the access to internal applications but not external applications (partial outage).
1.Both VPN and internet will have its own availability measures different technology even if they are network related. Best would be to divide and measure them.
2.E2E business KPI's is what customers are looking at.. as we have business processes.. we have to link infrastructure and software components associated and measure business KPI's rather than SLA
3.One of approach to calculate network availability can be weighted average between the availability of the Internet connection and the availability of the VPN tunnel: Network Network Availability = Weight external_apps x Network Availability Internet + Weight Internal_apps x Network Availability VPN
4.The two weights values should be set and discussed with the business architect considering business criticalness.
5.For example, let’s assume that 80% of the business applications used by the users are external (Internet) and that the remaining 20% are internal (VPN). If in one-year time interval, the Internet connection never failed, while the VPN tunnel was unavailable for 1 day, then the value of the overall network availability for that location is:
6.Network Availability = 80% * 100% + 20% * (99.726%) = 99.945%

So what you expert think about this? write your thoughts and concerns inthe comment section
Any guidance would be appreciated

User avatar
ITIL Expert
ITIL Expert
Posts: 129
Joined: Wed Mar 28, 2007 8:00 pm
Location: FRance

Thu Aug 23, 2018 10:07 am

Network availability as such has no meanings to me ... Network is a set of hardware and software components that are assembled to provide some services. What SLAs should measure is availability of services.

as such, in your scenario, I understand that "network" is contributing to 2 different services
- internet service: providing access to the outside world and resources , and possibly, allowing the outside world to exchange with you.
- internal (intranet) services: providing access to the internal world to allow users ans systems to communicate ans exchange information internally.

I would consider having one SLA for each service. If that has any meaning, calculate an average or, better, a weighted average depending on the criticality of each service for the operations of the company.

JP Gilles
User avatar
ITIL Expert
ITIL Expert
Posts: 3639
Joined: Fri Sep 15, 2006 8:00 pm
Location: London, UK

Fri Oct 26, 2018 4:46 am

To add to JP Giles point

There is a service here - 1 for remote site - to access the the HQ via the VPN Tunnel where the service provider is the Company IT Dept as the IT network people will manage the VPN Tunnel. The underlying circuits are managed from HQ to the various telcos under a UC. The SLA for the VPN Tunnel should take into account the effectiveness and efficiency of the telco as well as whether the underlying circuitry has redundancy. So if there are 10 remote offices; there are 10 SLAs for the Tunnel service and the requisite OLAs and UC for the internal teams that support or maintain the tunnels and the telco vendor for the circuitry.

There is a service for the Remote sites to the Internet where the Service Provider may NOT the company but the various telcos who provide the onnection. It depends on whom manages that

When you calculate Network availability - you have to do it to the service in question. So if the HQ is in Paris and the remore office is in Caen; then, the Availability will be against the telco circuit that provides the vconnectivity, the network equipment at both ends that make the VPN and the VPN itself
John Hardesty
ITSM Manager's Certificate (Red Badge)

Change Management is POWER & CONTROL. /....evil laughter
Post Reply