What’s lurking in your closet?
When most people think of closets, they think of a room full of clothes and shoes – or brooms and dust bunnies.
But in the world of data, clouds, and IT, a “closet” is a miniature server center: a room that houses working data processing and storage equipment. As suggested by their very name, server “closets” are often no larger than a small room with a locking door – and often the space was not designed to function as a server room. According to industry research, 17% of all servers are located in such closets – which represent more than half (52%) of all server facilities.
Inside these closets, however, the same dynamics are at work as they are in the largest data centers in the world: data streams in and out, operating systems run 24/7/365, cabling runs here and there, all using uninterrupted power, generating a constant hum. The result: a lot of energy is consumed and a lot of heat is generated.
A little history
So, how did server closets come to be? From nearly the very beginning of modern computing, the movement of digital information has involved centralized hubs: in 1939, Bell Telephone Laboratories connected remotely to a CNC machine to demonstrate its capabilities.* By the 1960s, central databases were connected to remote workstations where operators could access needed information, such as making airline reservations. But those central databases were huge, requiring entire floors or facilities to house punch-card processing and tape-driven systems.
As technology evolved, those early systems (ever hear of UNIVAC?) were replaced by semiconductors and silicon chips that allowed for the miniaturization of equipment, the storage of more data in the same space, which led to the creation server “closets.”
Today, a server closet is generally a set-aside room in an office or facility with its own thermostat, back-up power system, and integrated controller for use by a technician. Inside, there is often a metal rack with a handful of dedicated computer equipment acting as servers controlling the ebb, flow, and storage of information for the facility. The typical size of a server closet is usually less than 200 square feet, with server rooms and data centers much larger.
Regardless of size, be it a small closet or a cavernous data center, when you put computers together in one place, there’s a lot of generated heat that needs to be dissipated or at least mitigated – which is something I’ve been dealing with for more than 20 years as a computer and energy engineer. Although server rooms and closets don’t have quite the same infrastructure or require the same level of cooling and power requirements compared to data centers and co-location facilities, the principles are the same.
In today’s computing world, Moore’s Law** is becoming less of a driving force in computing economics. Why? Because the chip-making industry is no longer able to keep costs flat. As a result, operators of small data rooms and closets look at energy saving technology upgrades, such as:
- Energy-efficient servers (such as ENERGY STAR®-rated)
- Data storage management systems
- Server virtualization and consolidation using centralized or “Cloud” services
- Cost-effective HVAC upgrades and other steps that can save energy, save money – and other steps that can reduce or control costs
The challenge for utilities, then, is to understand that a data center manager’s main objective is and will always be uptime and data security; s/he may not have the time or the expertise to investigate detailed energy efficiency options. So to help these smaller facilities implement energy-efficient measures, one option is for utilities to develop prescriptive incentive measures that are simple to understand, evaluate, apply for, and implement. By developing a list of prescriptive measures that are designed specifically for the small data center customer segment, this should help move this segment closer towards achieving their goal of energy efficiency.
In the end, it is up to the utilities to provide their computing customers with easy-to-understand prescriptive measures with an attractive ROI. Anything else simply does not compute.
View the first two blogs in the series on data centers by John Greco: Why the End of Moore’s Law will Drive Energy Efficient Data Centers and Data Centers: What’s Behind Their Growth and Challenges They Face?.
DNV GL has successfully designed and implemented data center energy efficiency (EE) projects on behalf of our utility clients for more than eight years. Our team has significant experience on both new construction and retrofit data center EE projects from small server closets to co-lo facilities. We work through the entire cycle of project identification, justification and evaluation, and follow this up with detailed engineering analysis and post-installation review, assuring that the promise of a Data Center Energy Efficiency project is fully realized. Our team also is available to work directly with Enterprise Wide Data Center Operators to assist them in company-wide energy efficiency initiatives. For more information, contact John J. Greco or visit our Knowledge hub.
Mr Greco is a D.O.E. Certified Data Center Energy Practitioner, Certified Energy Manager, Certified Measurement and Verification Professional, Certified AEE Building Sustainable Energy Technician Trainer, Certified Lighting Energy Technician, and holds a Bachelor of Science degree in Electrical Engineering Technology from the State College of New York.
*Computer History Museum, 2016
** Moore’s Law states that IT equipment capability would double approximately every two years