Change may be constant in the data center, but tracking changes in the infrastructure
ITIL adoption in the data center
There's anecdotal evidence that ITIL is striking a chord among data centers. At Gartner's Data Center Conference in November 2006, 40% of 171 attendees said they have been implementing ITIL for less than two years, while 32% said they had plans to implement ITIL within the next 18 months. (Only 16% said they had no plans to implement ITIL).
Why the current popularity of ITIL given that the concept has been around since the 1980s? Sid Finehirsh, CEO of The CMX Group, an IT service management consulting company in New York, says that IT is increasingly being called upon to demonstrate its worth to the business. "IT is really about delivering information services rather than processing power or storage," Finehirsh says. "There's an increased awareness that IT is an expense item that exists to serve the business." ITIL, with its emphasis on standard processes and common IT definitions, is seen by many IT professionals as the means to achieve that service orientation, and do so in a way that the business can understand.
CMDB software helps facilitate ITIL, common language
So what precisely is ITIL? It's not a discrete technology although some software—notably systems management and configuration management databases (CMDB) – can support the adoption of ITIL by automating systems data collection and the mapping of applications to infrastructure components. At Grand Hilton Vacations Co., for example, a CMDB from Altiris enables the IT group to link IT assets with the services they deliver. ITIL, as its moniker suggests, is a group of books (also available on CD-ROM) that provide both a common glossary of IT terms and a standard set of IT processes that—once implemented—promise to introduce consistency across IT.
"ITIL provides the basic discipline to run IT professionally—with a common language and common processes," says David Pultorak, CEO of Pultorak & Associates, an IT service management consulting and training firm in Seattle. The underlying premise of ITIL: Without consistency and repeatable processes, IT can't effectively improve service levels.
The official genesis of ITIL goes back to a government office in the United Kingdom that sought to standardize on best practices for running IT in the 1980s. Over the years, the IT specific framework evolved to include eight core areas:
- Service support
- Service delivery
- Planning to implement service management
- Security management
- Information and communications technology (ICT) infrastructure management
- The business perspective
- Application management
- Software asset management
For the data center, the most relevant of the eight ITIL areas are service support and service delivery, in which processes for handling problem management, incident management, change management and release management are among the topics outlined.
Other popular frameworks, including the Control Objective for Information and Related Technology (COBIT) and the Capability Maturity Model Integration (CMMI) are broader in scope; which include processes for control and security, and development respectively.
What separates ITIL from other frameworks is that its development and periodic revisions are carried out by a vendor-neutral body of experts and practitioners committed to promoting best IT practices. In addition, ITIL is exclusively geared toward improving IT service management. "ITIL has more structure," says Finehirsh. "ITIL has an organization that supports it [The Office of Government Commerce in the U.K.] and defines what it is, and there are local user groups."
Customize frameworks to maximize ITIL effectiveness
Yet ITIL isn't a replacement for other frameworks, nor is it easy to compare and contrast the proposed benefits of the various frameworks. Indeed, some consultants say that it's perfectly acceptable to borrow bits and pieces of various frameworks and combine them in a manner that works best for your IT shop. Says IT consultant Dave Agonis, "process frameworks such as Six Sigma, CMMI and ITIL can coexist because no one framework is a silver bullet. It's okay to try a combination of processes in a blend to get the best capabilities."
Mike Phillips, the CMMI Program Manager at Carnegie-Mellon University's Software Engineering Institute (the developer of CMMI), agrees. "CMMI sets out goals—such as establishing a baseline of identified work products for configuration management—that can be satisfied by using ITIL processes—such as establishing a CMDB with access control."
Standardization foundational to ITIL
The concept behind ITIL isn't groundbreaking. "In the data center world, standardization is really important," says Bob Boyd, a contributing research analyst at Computer Economics, an IT research and advisory firm based in Irvine, Calif. Back when IBM effectively owned the data center, "standards were de facto…they were whatever IBM said they were."
With ITIL's common glossary of predefined terms, IT can eliminate communication gaps that often result simply because various groups use terms such as "problems" and "incidents" interchangeably, for example. In ITIL parlance, an incident is any nonstandard operations event that could cause an interruption to, or reduction in the quality of service. A problem, on the other hand, is an underlying cause of one or more incidents. .
At office supply company Corporate Express based in Broomfield, Colo., data center manager Dave Hines is looking to ITIL to establish processes in order to provide structure. "We'll start by setting up processes that we don't even have yet," he says. "When new equipment comes in, we don't have a process for notifying the data center; there will be a new server, and we don't know what it is, where it's going or what kind of connection it requires," he says.
As a framework covering the range of IT services, implementing ITIL can seem overwhelming. Consultants and practitioners say the best way to begin is to start small. Focus on one or two of IT's most vexing pain points. For data centers, often the biggest headaches are in the areas of problem, change and configuration management; those areas are good starting points. In terms of service management, start with the areas that cause the most disruption to end users. "If there are outages or issues of low availability, those are things that are most visible to your end users," Agonis says. "If that's where your problems are, something is wrong with your change management processes."