The CMDB in 2008 – It’s a moving target and many will fail to hit the mark

Updated 4 June 2008

Now that most of the vendor community has come to terms with the ITIL v3, the new offerings to the general market are starting to emerge. Some are obviously just re-branded and tweaked with the V3 rinse and others appear to be a consolidated effort to encompass all of the requirements of service management life cycle concepts as detailed in the ITIL v3 books of Service Strategy, Service Design, Service Transition, Service Operation and Continual Service Improvement.

ITIL v2 was highly criticised for its view on what constitutes an ITIL compliant CMDB, and the reverberation of the impact will be felt for years to come. The impending train wreck of a failed strategic vision is inevitable for some organisations. Multiple streams of understanding now exist in the industry and it is highly likely that a consolidated view point will never be achieved.

Let’s take an example of implementing a top tier service management lifecycle solution including the holistic CMDB within a large organisation. These vendors would mandate that to gain the stated benefits of their offering, organisations must migrate their existing Change / Incident / Capacity management processes (etc, etc) to the new platform. This maybe a fair assumption for some, but in many cases the organisation just wanted to achieve the implementation of a CMDB in order to support their overall move to an ITIL based framework. This presents both considerable cost and risk to the overall operability of the IT environment just so IT Management can put a tick in the box and state that they have now achieved a perceived level of maturity under the framework model. Both from a costs perspective but also from a business and systems risk management point of view there are easier and more effective ways to get the job done.

It now needs to be accepted that organisations of various sizes may actually require a variety of information subsets and those may not be common to all, and that the one size fits all approach to a CMDB is fundamentally flawed. As well as the purist view of the requirement of a single federated data source.

At a recent seminar there were cases in organisations of less than 300 people wanting to scope the implementation of a CMDB when the change process is at a low maturity and not supported by all stakeholders in the organisation. Clearly their technical teams were driving this effort and wanting output and controls from the CMDB that were very different to senior management and finance group. The service manager wanted SLA measurement and capacity information, but the change management guys were happy with their spreadsheets. Clearly a vendor driven end to end service management software platform is not the answer, but a solution to specific problems is still required.

The pressure is certainly getting turned up on the business front; CEO’s, CFO’s and CIO’s alike are directing that the organisation strive forward with a “Do More with less” and “work smarter not harder” agenda. Great in terms of bottom line until someone wakes up to the fact that the skills base has been eroded only to be replaced by financially motivated vendor proposals for every new solution.

Amid the current global economic climate, business is demanding more of its resource base with the theme of “Doing More with Less” and “Work Smarter not Harder”, so its time to get back to some basics. “What are the true business requirements for the implementation of this type of service management structure?”

In a large organisation the one minute manager rarely has the capacity to delve into the micro view of the IT governance framework, but the catch cry can be heard through the hallways, Give me a robust governance and service management best practice framework which suits our organisation. Typically this just ends up as a few key staff going out for training with the result of a proposal for an "Off The Shelf" framework which gets implemented with religious zeal.

Modularity must be raised as one of the key guiding principles in this debate as well as a healthy level of pragmatism and skepticism. Understanding exactly what it is, that the organisation needs to achieve is a fundamental starting point. It maybe things such as understanding Data Centre capacity constraints, risks to service continuity or SLA’s associated with scheduled changes to key infrastructure components, building a database of assets from automated network discovery and the ability to build a baseline of assets, software inventory, discovery of unauthorised changes or the ability to have a known state of configuration of specific devices.

Rather than following a prescribed Service Management Framework there needs to be some exploration and definition of the uses for a CMDB. Don’t buy a Basenji if you want your dog to bark. Don’t get a Great Dane if you’re not prepared to shovel manure twice a day, every day.

Some of the key services and required outputs of a CMDB have been highlighted as:

Configuration Controls Automated Network Discovery Asset and Software Management Up to date risk management assessment capabilities Data Centre Capacity queries Data Centre / Server Room cable and power management Service to systems mapping Application and services, service level measurement

Below are some links which may assist in finding the right solution for your organisation:

Configuration Controls

Tripwire Enterprise

Open Source Tripwire

Puppet

Automated Network Discovery This site lists a few well known packages for network discovery and mapping

NetworkIntrusion.co.uk

Asset and Software Management

OCS, It's an open source product and a personal favourite

Data Cabling

Some pics of how cabling should look on day one

Service Level Measurement

Multiple references for Service Level Measurement with Nagios

Data Centre and Enterprise Service Configuration

Check out the AssetGen Suite of Management Solutions --- This really has to be noted as a benchmark for visibility of systems and services in the enterprise

Please don’t forget about the great features of the Microsoft Active Directory and the Microsoft view of Service Management Functions

Some great open source software is also available already configured under a VM Guest download the ECDB Toolset from CMDB.info at the downloads page

ECDB Toolset from CMDB.info

Just think about the scenario of preparing a business case to present to the CFO for funding approval. Software has been evaluated, existing service management processes have been mapped and undergone a maturity assessment. The justification statement highlights the fact that the organisation is moving to a fully compliant ITIL framework and now requires funding (including resourcing) of up to $1m+ over the next 12 months to achieve this goal. Don’t expect the CFO in today’s financially constrained world to be sympathetic to achieving this divine level of best practice management and process maturity without exploring all of the options, regardless of the size of your IT environment. A multitude of strategic solutions are available in both open source and proprietary solutions to achieve specific goals.

As computing technologies advance over the next few years, we will see a metamorphosis of this evolving market segment which will change the way we all view the enterprise computing model and the application of structured service management concepts.

Feel free to provide some feedback if you feel inclined. (email addresses are on the contacts page)

Regards

Brendan Martin