Problem solve Get help with specific problems with your technologies, process and projects.

Using ITIL to bridge the IT-facilities gap, boost data center efficiency

True data center energy efficiency can't become a reality until the IT and facilities departments communicate effectively about available data center resources. ITIL best practices provide a framework that can help bridge the IT-facilities gap and ensure that all the key players are part of the energy-efficiency conversation.

At first blush, it may seem absurd to use the IT Infrastructure Library (ITIL) and energy efficiency together in the same sentence. After all, ITIL is a set of IT best practices governing, among other things, change, problem, asset and capacity management. It provides a systematic way to assure availability and reliability as IT rolls out new applications, does maintenance, installs severs, virtualizes, and reacts to problems and threats.

Seen from another perspective, ITIL is also a doorway to IT culture. It provides a language that the employees in the IT shop use and can relate to. That's where the opportunity arises. If the ITIL language and framework can be applied to all facets of the data center on both the IT and the facilities sides, it can help both sides of the data center to communicate and shrink the IT-facilities divide.

IT is focused on maintenance and rolling out new applications, and ensuring that changes and problems don't disrupt service. As a result, IT may not be aware of reaching capacity limits in power, cooling or space until it is too late or until something goes wrong. On the other hand, the facilities group is typically very aware of the IT "burn rate" when it comes to power, space and cooling.

For an organization with non-communicative IT and facilities departments, it is common for IT to blame the facilities team when a data center runs up against the limits of its available power supply. And facilities often shoots back that IT never bothered to ask. How can true efficiencies be achieved if the two groups responsible for operations don't communicate about the most critical resources for data center operations?

The IT-facilities divide has been pointed to repeatedly as a barrier to more efficient data center operations. The McKinsey Report: Revolutionizing Data Center Efficiency, for example, claimed that the data center efficiency problem would not be solved until facilities was placed under the command of the CIO, making the data center power bill the direct responsibility of the IT department. Currently, the cost of the power consumed by the IT equipment is not at the top of the CIO's list of priorities.

ITIL best practices provide a framework for reducing the IT-facilities divide, facilitating communication and unifying efforts by both groups so that efficiencies can be identified and implemented. The tools exist within the ITIL disciplines. We'll focus specifically on the disciplines of change management and capacity management.

ITIL-based change management and facilities
IT departments live and die by their change management processes. If these processes are not exacting and well run, the resulting collisions and downtime may be career ending for IT managers.

It turns out that change management is used by facilities organizations as well. Within the framework of ITIL change management best practices, this shared expertise can yield significant rewards. For example, most IT departments regularly hold change meetings to make sure that proposed changes will not cause failures or downtime. Facilities is typically not represented at these meetings. But imagine the advantage if attendance by facilities personnel was required. Here are some of the things that facilities could bring to the table.

  • Will proposed changes overlap with maintenance on UPS (uninterruptible power supply) devices, backup generators or cooling equipment, creating unacceptable risk?
  • Will there be enough floor space and power for proposed installations? Will the strategic location on the floor relative to power whips and cooling capacity be affected?
  • At the current burn rate, when will IT run out of floor space, cooling capacity or power for additional installations?

These practical change management issues can only be communicated by means of shared language and clear metrics. Facilities personnel are in the position to use the shared language to supply important but often missing metrics.

  • Facilities maintenance schedules
  • Available power, space and cooling
  • IT usage rates (burn rates) for power, space and cooling
  • Projections of capacity limits at current burn rates.
  • Costs associated with current usage rates for power, space and cooling (total cost of ownership)

Such metrics not only make for a clearer, more complete understanding of change issues, they also provide a much more comprehensive view of data center operations and costs.

ITIL-based capacity management and facilities
Capacity management is the ITIL discipline that focuses IT on proactively ensuring that future business requirements for computing are met cost effectively. It involves understanding usage trends and analyzing likely changes in demand and services as well as in computing power, memory, networking and storage. Capacity management also involves planning for and justifying future investment in capacity expansion.

On the IT side, capacity is usually thought of in terms of throughput, transactions per minute, IT load and so on. How much can the IT equipment do and how quickly? How can systems effectively migrate to new applications in order to add new services? How can IT install new numbers of servers -- perhaps at a greater density -- to maintain the required level of service? Is there a cost savings to be gained by moving from hardware by manufacturer X to hardware by manufacturer Y? All of this IT capacity, often referred to as IT infrastructure, requires a sufficient amount of accompanying critical facilities infrastructure, in the form of the data center holy trinity: power, cooling and space.

Here is another place where communication between IT and facilities can provide a more complete picture of data center operations and identify opportunities for efficiency. To put it crudely, facilities delivers the kilowatts that IT needs. Facilities manages the mechanical delivery systems in the data center and, as a result, has critical information about data center capacity that directly affects the capacity of the IT equipment. Will moving from hardware X to hardware Y increase IT's power burn rate, shortening the distance to the data center's power-capacity horizon? Will the new servers even fit in the racks? Once again, the participation of facilities in the conversation can produce substantial benefits.

Practical steps for facilities teams to get onboard with ITIL
For most organizations under the current model of data center operations, initiatives by the facilities unit is the most practical method for achieving facilities-IT communication via ITIL best practices. By crafting their communications (concerns, information and metrics) with reliability and availability in mind, facilities personnel can bring a certain degree of enlightenment to their IT peers.

For example, most data centers don't have the capability to bring in more power. It may be possible to bring in the necessary transformers, but in all likelihood, the physical space to upgrade the UPS isn't available. Ten years ago, a 1,000-square-foot data center required about 300 square feet for the UPS room. (This is scalable to about a 10,000-square-foot data center.) By the late 1990s, the increased density on the data center floor required a mechanical room roughly 50% of the square footage of the data center. Currently, the mechanical, electrical, plumbing (MEP) room in a data center requires about the same amount of space as the computer room.

Facilities personnel know the requirements of these spaces. They also know what kind of output can be expected from the equipment in these spaces. They are aware of when the capacity of this equipment and space for additional equipment will be maxed. This is not information that IT is likely to ask for, yet without it, the operating parameters available to IT can't be understood. Unless facilities takes the initiative to proactively communicate this information to IT, the management of the data center operation becomes like the proverbial blind men with the elephant: No one knows exactly what he's dealing with.

What can facilities personnel do to insert themselves into the conversation? Here is a list of practical steps.

  • Learn the main business applications (POS, online sales, reservations, etc.) that IT supports and where facilities support work intersects them.
  • Start a process of weekly communication with IT using ITIL terminology.
  • Participate in the change meeting and contribute your schedule of expected changes to help avoid collisions.
  • Create metrics that convert facilities services into availability and reliability measures (more on this below).
  • Find out who is in charge of IT's capacity planning and how to engage with that person.
  • Use ITIL best practices to create common ground. (Change management and capacity management are two likely disciplines, and energy-efficiency practices can be inserted into each).
  • Bring information on data center capacity (IT's burn rate for space, energy and cooling versus available capacity versus limits to expansion) to change management processes. Find out how to contribute to the change management database.

Facilities has ready access to metrics that are extremely relevant to IT. For example, by creating a facilities scorecard that quantifies its performance, facilities can communicate its contribution to overall data center availability and reliability. The scorecard can cover the following areas.

1. Finance

  • Year to date expenses verus previous
  • Monthly usage
  • UPS/cooling costs
  • Costs of specific usage

2. Process

  • Number of Severity 1 problem tickets
  • Open root cause analyses (specifying area and date (e.g., UPS #2, expected completion date 6/1/09)
  • Updated capacity charts including IT burn rate

3. Performance

  • Mean time to repair (MTTR)
  • Mean time between failures
  • Availability
  • Saves and near misses, utility failure, pump failure, weather event)

4. Projects/Plans

  • UPS maintenance (with specific dates)
  • uCode updates (with specific dates)
  • AC quarterly maintenance
  • Scheduled generator tests (with specific dates)

All of the information conveyed on such a scorecard has direct relevance to IT, will increase IT's knowledge and boost the likelihood of further communication. The information also provides opportunities for discussion of possible efficiencies that can extend capacity and help the bottom line.

ABOUT THE AUTHOR: Dave Rosenberg researches and writes on IT, critical facilities and high-density computing issues with a focus on sustainability. He formerly served as Content and Editorial Director for the Uptime Institute.

What did you think of this feature? Write to's Matt Stansberry about your data center concerns at

Dig Deeper on IT compliance and governance strategies

Start the conversation

Send me notifications when other members comment.

Please create a username to comment.