“Developments around the CMDB are going to drive the IT roadmap for some years to come,” predicted Siki Giunta. Giunta was Managed Objects’ CEO prior to October 2008, when it was acquired by Novell, where Giunta is now VP strategic business. A rather sweeping statement perhaps, but does Giunta’s prediction  stand up to scrutiny? Is the configuration management database, or CMDB, still an important weapon in the CIO’s arsenal, or was it over-hyped, over-complex and ultimately, rarely properly implemented?

It’s certainly true that many companies have bought into the idea of business service management (BSM) – an approach that advanced systems management by enabling it to be tied into the customer experience and service level agreements. BSM closed the divide between the availability of the IT infrastructure and the real performance that it was delivering to the business and its customers.

A configuration management database (CMDB) is a repository of information related to all the components of an information system. Although repositories similar to CMDBs have been used by IT departments for many years, the term CMDB stems from the IT Infrastructure Library (ITIL).

“Business service management (BSM) is on the rise,” says Thomas Mendel, VP and research director at Forrester Research. “The CMDB is one of the key components of a BSM implementation. The CMDB has come to the forefront of IT managers’ attention through the combination of new technologies such as application dependency mapping and the quest for management process improvement (ITIL).”

In the ITIL context, a CMDB represents the authorized configuration of the main components of the IT environment. A key goal of a CMDB is to help an organization understand the relationships between these components and track their configuration.

 

Process of configuration

The CMDB is a fundamental component of the ITIL framework’s Configuration Management process. CMDB implementations often involve integration with other systems, such as Asset Management systems. These integrations may make use of either a real-time, federated design or an extract, transform and load (ETL) approach.

But just because the CMDB was conceived as an important element within ITIL, that doesn’t make the CMDB pivotal to a BSM strategy. That’s because it’s possible to go down the BSM route without following all of the rigour of ITIL – indeed many companies are doing BSM on their own terms without paying much heed to ITIL, or perhaps attempting what you might call ‘ITIL light’. They are guided by some of its principles but are not yet ready to commit resources to the discipline of a full-blown ITIL strategy.

But for those attempting to follow ITIL more closely or indeed in its entirety, the ITIL approach fosters an approach to IT service management that is largely based on the development of an understanding between business services and the IT elements drawn on to deliver them.

A first stop on that journey is the construction of a configuration management database or CMDB, as a repository of asset and service data. All the big systems management software suppliers have now delivered products designed to support CMDB initiatives, while niche vendors like Managed Objects (acquired by Novell) and Tideway designed service management applications to feed data into a CMDB or help visualise and analyse data once in a CMDB.

 “We see the configuration management database as sitting at the very base of all analysis and decision-making that IT departments will be making in the future,” says Giunta. “What is my overall IT capacity looking like? How much does this service cost to deliver? Where are the critical elements of the business support infrastructure? Should I virtualise this part of the application infrastructure? These are the sort of questions and decisions that the IT department is starting to make about the nature, the cost and the quality of IT services. And they will all rely on the foundation of a CMDB.”

 

The ideal CMDB

The ideal CMDB, should, according to ITIL, provide accurate information on configurations and their documentation in a way that will support all the other service management processes of incident, problem, and change management. Analyst houses and software vendors alike are pushing the idea of a federated or virtual CMDB as one means of reducing the complexity of deployment, using software tools that automate topology discovery to accelerate the CMDB build process. That requirement has forced a good amount of market consolidation.

BMC and CA developed CMDB and discovery systems for themselves, but as Michael Allen, Compuware director of service management observes, “There’s been a real land-grab by the other big systems vendors, who all want to get onto this area. IBM acquired Collation, EMC bought n-Layers, Symantec took over Relicore, and Mercury had Appilog which of course are now both owned by HP.” Late last year, Novell bought Managed Objects, which had been something of a poster child of the CMDB market.

When IBM bought Collation back in November 2005, it explained the move by saying that Collation software automatically captures information about IT resources, such as servers, applications and databases, and displays it on a detailed map. This enables IT to better understand the impact of changes to an IT environment — such as how a security patch can trigger a ‘domino effect’ of unexpected problems that can bring down an online business. It also helps shrink the time it takes to discover and correct a problem, according to IBM.

“Without a solid understanding of how different pieces of the IT puzzle fit together, a single change can slow down or halt an entire IT system,” said Al Zollar, general manager, IBM Tivoli Software. “Collation software allows companies to visualize different change scenarios so they can analyze the implications before problems occur.”

Collation software is said to help IT professionals manage the availability of their systems when making changes, which helps ensure that applications remain up-and-running around the clock. IT professionals can also see how technology supports business processes such as order entry, supply chain and enterprise resource management.

Using IBM Tivoli systems management software, customers can take the information captured by Collation software and model what will happen when a change is made. They can also test the environment for unauthorized changes and share a common view of how different pieces of the IT infrastructure work together to deliver IT services. This helps companies better manage their software deployments, making them more responsive to new business requirements by shortening the time it takes to release new applications or update existing ones.

Today IBM calls its CMDB the Tivoli Change and Configuration Management Database (CCMDB) – “a platform for storing deep, standardized enterprise data.” The product is said to automate data, workflows and policies, to align IT infrastructure management with business priorities.

IBM draws a clear distinction between a database that stores configuration data – a configuration database of CDB – and a configuration management database or CMDB, which crucially adds the element of management.

Most of the vendors today see a need for software that will automatically populate and update the CMDB data by discovering which part of which application or business process is running on which infrastructure components. From there IT can start to trace how the interdependencies between them map against each other. By adding visualisation and analytics to that process it becomes possible to show the impact and root cause of any change that occurs in the infrastructure.

Compuware does not ship a CMDB of its own but is able to integrate CMDB data into its Vantage Service Manager service management suite. Compuware notes that most organisations have several monitoring tools already in place. It espouses Vantage Service Manager as a way of bringing these together, with its capacity to integrate infrastructure management applications such as IBM Tivoli, BMC and HP; service desk software such as Remedy and Peregrine; and asset and configuration management systems, including any home-grown or commercial CMDB.

Allen argues that such a set-up delivers a system that is best suited to monitoring the end-to-end behaviour of an application and the state of the underlying, supporting IT infrastructures: “In many situations the degradation of service performance or complete failure of a service can be traced back to an IT change made on a production system. There is a high level of correlation between end-user problems and IT change.”

 

Rising tides

Richard Muirhead, chairman and CEO of Tideway — one of the few remaining pure-play application dependency vendors in this sector — maintains that IT executives are still getting to grips with the many ways the CMDB underpins and permeates various aspects of IT: “Once you start to understand the relationships between configuration items and can view the interconnections that exist in business processes, you start to see that it could revolutionise the use of existing service management toolsets.”

BMC, which has adopted BSM as an over-arching strategy for product development and deployment, as long ago as June 2006 introduced BMC Atrium 2.0 and a new set of integrated workflow solutions designed to automate links between what it calls ‘Routes to Value’ or RTVs.

The BMC Atrium 2.0 architecture introduced a second generation CMDB data model, a more robust reconciliation engine, an improved ability to accept data from a wider range of federated sources, an updated graphical interface and improved access to third-party tools and management information databases.

One of BMC’s customers is DealerTrack, a provider of on-demand software and data solutions for the US automotive retail industry. The firm says it is using BSM, including CMDB technologies, to help improve service delivery to its clients. “Business Service Management from BMC Software helps us keep our on-demand products and services available and running at peak performance,” explains Charles Giglia, SVP and CIO at DealerTrack. “Faster transmission of data between dealers and lenders helps them close deals in less time.”

Developing and deploying a configuration management database is a major undertaking, and the latest thinking is that the most effective approach is indeed the federated data model. This has configuration-item data held in a common data store with other less critical attribute data federated and fed from other application databases. That way, configuration data can be shared without the high set-up and maintenance costs.

Also, discovery is key, as outlined by Anthony Esposito, senior architect for an IBM CMDDB customer which is described as a major international consulting firm which provides guidance for Fortune 500 companies via over 80 offices worldwide: “Choosing automated discovery, of whatever form (in our case what is now IBM’s Tivoli Application Dependency Discovery Manager or TADDM), is of little use unless you have a CMDB in which to store what you have discovered,” says Esposito. “Plus, remember, discovery is something that you must run frequently (like a web bot) to find out what has changed. Choosing a CMDB is as important as choosing discovery…and vice versa.”

Esposito’s firm had quite a shock when it ran the IBM Tivoli discovery technology: “When we started with discovery, we found some surprises. For example, we found that the number of Linux servers and Windows servers was not quite what we had expected. We had thought we had anywhere between 500-600 servers in each of our data centres. It turned out we had about 800 of them, in each. We were a third bigger than expected. This was a shock, especially when we considered the implications for licensing and everything else downstream.”

“Our primary reason for using discovery, however, was to discover the dependencies between tools, networks and applications,” says Esposito. “There were plenty of surprises here as well.”

But according to Esposito, “Discovery and a CMDB are themselves of little value unless you have means and tools to analyze the information that has been discovered and then stored.”

CBR opinion

ITIL-driven CMDB projects tend to be ‘bottom-up’ – they start out as incident management or change management automation projects. But ultimately, CMDBs tend to add a new service layer, which needs its own visualisation and analytic capabilities for it to add real value.

Of course there is nothing new about auto discovery tools generating visual maps of the IT infrastructure, but BSM vendors are now able to provide tools that not only identify software applications and their interdependencies, but integrate well with change management. Together, the CMDB and related visualisation and analytic technologies have the potential to  bring much-needed business meaning to business service management (BSM) practices.

 

Also in this report: CBR looks at the adoption of ITIL and how ITIL version 3 has moved the game along, here. Also, we assess the general move from systems management to business service management, and ask what the business case is, here.