The demand for ever-more sophisticated business services and the migration towards on-demand, high availability, converged enterprise networks is creating considerable challenges for those responsible for ensuring 100 percent availability of IT services, and that in turn is driving the introduction of the ITIL best practice standards for IT Service Management.

At the core of ITIL’s framework are configuration, change and incident management processes, so the accuracy and relevance of the information within the main configuration management database (CMDB) is critical in delivering effective IT Service Management (ITSM) processes. It has been suggested that 80 percent of critical system outages are the result of people or process failure, the main reason being change. This is in turns leads to a cycle of failure - poor information is destined to fail, and the subsequent recovery process is hampered by originally poor configuration data.

The management of changes to the CMDB is critical, as service delivery managers must know the implications of carrying out a change before it takes place. Avoiding the ‘butterfly effect’ - a small change having wide-ranging and unanticipated effects - is essential if 100 percent service availability is to be maintained. As David Cuthbertson of the British Computing Society points out, “How can you manage IT services if you don't know what equipment you have, where it is, how it’s connected and the impact of changing it?”

The configuration management challenge
Configuration management is cited as the most difficult IT discipline to implement, because constant changes to the IT infrastructure and the lack of suitable tools to intuitively manage the processes mean that manual effort is needed to keep the CMDB accurate. Rather than a fully integrated CMDB addressing OSI layers 1 to 7, the majority of organisations rely on tribal knowledge and/or numerous sets of spreadsheets, diagrams, whiteboards, post-it notes and the knowledge in peoples' heads to document the physical infrastructure elements.

A number of software vendors have sought to overcome this need for manual CMDB maintenance by developing IT service configuration management applications that dynamically discover substantial amounts of interrelated network information.

Deploying such applications and processes upon a traditionally managed cabling infrastructure is well-nigh impossible, as the connectivity and asset information contained within manually-maintained tools and documentation processes is inevitably inaccurate or outdated, and therefore cannot be safely incorporated into an accurate CMDB. This severely limits the speed of application deployment and the quality of ITSM, and is a major barrier to successful ITIL adoption.

What is IIM?
One route around this is to implement Intelligent Infrastructure Management (IIM) tools. These combine a set of technologies to do the following:

  • Self discover network connectivity from end to end
  • Provide real-time IP asset management to the physical location
  • Generate alarms and events for unauthorised activity on the physical layer
  • Share information with various work streams and core applications

    Key components of an IIM solution are:

  • Relational infrastructure database
  • Patch panel sensing system that detects the state of patching and constantly monitors for change
  • IP discovery engine to add end devices and switches to connectivity circuits
  • Multiuser interface and third party application integration
  • Work order system

    As part of an ITSM strategy, IIM can enable IT managers to create an infrastructure platform capable of providing a 100 percent accurate, real-time, trusted source of connectivity and physical asset information, which can be incorporated within the core CMDB and subsequently used to enhance associated ITSM tools and process.

    IIM solutions such as iTRACS, iPATCH and RiT provide an auto-routing capability within their work order management functionality that automatically routes the required services to the desired servers via the most efficient and effective cable links. The work orders required to execute the activity are automatically created, issued to the appropriate technician and managed by the IIM system. Any actions that do not adhere 100 percent to the work order are identified and raised as unauthorised actions requiring attention.

    Using IIM solutions to automatically provision services within a standard office environment should improve change efficiency (and hence reduce cost) by a minimum of 40 percent. Within the data centre, IIM solutions have been proven to reduce server commissioning time by up to 80 percent, whilst simultaneously reducing the number of incidents caused by poorly executed change. If incidents do occur, IIM solutions can reduce mean time to resolution (MTTR) by up to 45 percent.

    Terry Riches is a senior business manager at IIM implementation specialist Comunica.