Assured and automated

The term service assurance has been around for a long time in the telecommunications industry, much longer than in IT.

With the service awareness created by the Information Technology Infrastructure Library (ITIL) over the last years, businesses now demand that IT assures the availability and optimal performance of services being consumed by the business or end-user.

All of this needs to take place in accordance with specific service level objectives supporting specific underpinning contracts and service level agreements. Hence, as far as the business or the end-user is concerned, it is - 'performance first'.

The service stack is a term used to describe the infrastructure supporting a specific service transaction path. It makes it possible for end-users to transact from an end-user device across a network, where the end-user is executing a specific transaction that typically communicates to a system or database located somewhere in a data centre - as depicted below.

This service stack can be made up of a combination of physical or virtual components or configuration items (CIs).

From the end-user perspective, it is all about the perceived performance or lack of it. When the end-user says the service or application is slow, it is slow.

Typically, there is little concern about what is causing it, hence - performance first. It is the service provider or IT in an internally focused environment that needs to ensure the optimal experience.

Many organisations still operate in a silo manner, where there are specific IT divisions or departments each focusing on their own technology, and they don't always communicate very well with one another - the end-user is the one that suffers.

When the end-user is complaining about a slow service, there might be the following scenario:

* Network division checks - everything is up, must be the server or the database.
* Server/system division - everything is up, must be the network or the database.
* Database division - the database is performing well, must be the network or the server.

Then why is the end-user still complaining about slow service?

Discovery

Users need to:

1. Discover the infrastructure with their specific upstream and downstream relationships. They must understand how the infrastructure supports the applications and the transaction flow across it.
2. Discover all of the applications and how these are dependent on specific infrastructure.
3. Understand the interdependencies between infrastructure and applications and how they support services consumed by the business or the end-user.

Service assurance

Now that users understand the interdependencies in the service stack, they need to monitor the performance of all of the pieces or CIs in the service stack in relation to one another.

The following approaches are necessary:

* Monitoring the entire physical and virtual network; server/system components for performance and availability.
* Monitoring the application or transaction flow across the various CIs, plus understanding the specific touch points.
* Monitoring the transaction relationship to the network, server and application interdependencies.
* Monitoring the application performance in relation to its dependency on the various CIs.
* Monitoring the end-user experience in conjunction with the application performance and flow across the CIs.

From the end-user perspective, it is all about the perceived performance or lack of it.

Hannes Lategan is solution strategist at CA Southern Africa.

Because of the type of management solutions needed to perform the functions described above - it still makes sense to have different IT divisions in an organisation, each with their respective subject matter experts.

However, it is important that the information gathered from the various service assurance tools be consolidated into a service operations view that spans all of the silos.

The 'service operations' solution will then provide both a 'business service' topology and a 'business service console', from where users can provide the right information to the right people:

1. Executives and potentially customers, as well as service managers, will have insight into the quality health and risk associated with each service they consume.
2. Operations staff in the operations centre can have insight into a business service topology spanning multiple silos, with drilldown capability to specific domain residing 'service assurance' tools.
3. The SME per silo can still work within their respective silos, utilising the specific technology relevant tools they need.

The secret is to understand what the organisation's purpose is:

* What does it make money from?
* What services are consumed by the business and the end-user?
* How do these services need to perform?
* What infrastructure supports them?
* How do I monitor them?

Then contact a company that can be considered as an IT management expert in its field to help get value from constructing such a service assurance solution, which will cater for the dynamic environment consisting of physical, virtual and cloud solutions.