Monday, February 13, 2006

Analytics & IT Service Management

I saw a number of recent articles and webinars using the term 'analytics. One of them, 'Analytics' buzzword needs careful definition, by By Jeremy Kirk, IDG News Service, 02/13/06 you may have seen in the recent NetworkWorld Newsletter.

In this article, Gartner is quoted as defining analytics as:

["Analytics leverage data in a particular functional process (or application) to enable context-specific insight that is actionable." It can be used in many industries in real-time data processing situations to allow for faster business decisions].

In an ITIL-based context analytics can be viewed in many ways, since ITIL defines multiple processes, but the time savings associated with implementing IT service management is largely in the diagnosis and isolation of problems (IDC estimated as high as 75%). This is particularly true for n-tier infrastructures which are driving the need for ITSM in the first place.

In this case, IT service management analytics must include the ability to correlate real time data from every layer of every component of your n-tier infrastructure, automatically isolate the root cause AND present that in a way that is easily understood and actionable (to use Gartner's term).

This capability benefits many ITIL processes. The Service Desk, Incident and Problem management are obviously improved through a reduction in incidents and much greater ability to achieve proactive problem management. Service Level Management obtains a clear view of all interdependencies and real time and historical information on service performance. Capacity and Release Management are able to quickly isolate potential performance problems BEFORE changes are implemented (which, by the way, benefits both Change and Configuration Management).

When analytics is implemented in a way that multiple orgaizational silos can leverage the information in a way that's meaningful to them, even Application and ICT management benefit. Each obtain an easy to use, visual representation of each IT service which can help them understand and shift paradigms to a service orientation. In the case of ICT management, event filtering is highly automated which can lead to a significant improvement in the utilization of staff.

The key to effective monitoring of n-tier infrastructures is very much about analytics, but this does not mean men in white coats need to be crunching numbers, setting rules, etc. Successful anaytics in monitoring n-tier infrastrcutures is about how quickly and easily can the appropriate stakeholder obtain actionable information about what's happening with the service(s) from whatever perspective is relevant to them. Without taking months, without writing rules, and in a way that can keep pace with the business.

Silo-ed analytics may simply make things worse, since greater costs are added and much greater integration effort to achieve end-to-end services analytics. These silos can take different forms as well, such as response time 'analytics', or network 'analytics', even data base or application 'analytics'. In an IT service management approach, analytics should happen at the service level.

So, go ahead and leverage analytics for your environment. But understand why you're crunching the numbers in the first place.