INDUSTRY watchers have long complained about IT management as we know it. They predicted the dominance of high-tech silos, futile finger-pointing sessions and poor application performance would cause breakdowns in troubleshooting and multiple escalations of user complaints.
While IT operations managers want to focus on improving services to the end user or customer, technologies such as virtualisation, cloud computing and mobility are now making it imperative to adopt a service assurance strategy even more quickly.
Here are five simple steps that move IT managers away from frustrating, ad hoc IT management approaches to more efficient operations, optimised application performance and satisfied end users and customers:
1. Change minds to advance IT
As with most IT projects, the initial step involves a change in thinking or a cultural shift. For many in IT, the outcome previously could have been measured by server response time or network uptime statistics, depending on the domain. In today's dynamic environment, the outcome must be measured by the end-user (whether they are internal or external customers) experience with the delivered IT service.
That means IT needs to monitor the individual components comprising the overall service as it always has done, but also focus the attention on how those elements support the service and how that service fulfils the customer need.
2. Integrate existing IT tools
The next step is integrating the service management lifecycle and the monitoring tools perspective. Bringing together what have historically been separate tools and aligning them as relevant to service objectives is a big challenge technologically. Not all toolsets can easily make the transition to the service assurance model when IT organisations make this choice.
Businesses need more than a dashboard with a red/green light to indicate if there is a problem or not. IT operations will have to identify services and model those services in such a way that any changes to the underlying application and infrastructure components supporting the services are automatically updated in near real-time.
This ensures that operations can easily track down the source or problems and reduce downtime as well as improve the end-user experience and bottom-line results.
3. Prioritise remediation
Once an organisation understands who is affected and how they are affected by a problem, it can begin to work on solutions. A customer may be willing to wait a reasonable amount of time to update his contact information, whereas a login process that takes the same amount of time could well influence the same person to look for another provider. The end goal is to connect the end-user transaction experience to business outcomes.
4. Resources, just right
At the same time, an effective service assurance model allows an organisation to view and map transactions effectively so that there is a better understanding of IT and business requirements. An enterprise can then assign the optimal combination of resources - hardware, bandwidth and network capacity - to applications rather than over- or under-provisioning resources.
An international food manufacturing company has accepted that service assurance ties neatly into global operations. The company has a goal of doubling growth every seven years, and thus needs to have a good grasp of everything from transactional activity to network performance across geographic regions. "Clearly, we don't want to overbuy capacity, infrastructure and bandwidth that we don't need," explains the IT team lead. "The goal is to anticipate capacity and do predictive buying so we're not late adding systems."
5. Fine-tune, then fine-tune again
Adopting a service assurance approach to IT management won't resemble a one-off IT project. It will be an on-going process that delivers incremental returns along the way, and IT operations managers must decide from where they want to begin. If the ultimate desired result is an improved end-user experience, IT departments must define what that means to their organisation.
Once the transformation has occurred, IT organisations could start to consider metrics such as, "How much money is lost if we make this infrastructure change?" or "How does this operations incident put my SLA compliance at risk?"
It is too easy for customers to switch to another service, with another service provider; IT needs to be able to identify the metrics that mean the most to their business and ensure that they minimise operational as well as reputational risks.
(Stephen Miles is vice-president of service assurance for Asia Pacific & Japan at CA Technologies.)