ebm-papst Fans, Blowers and Technology

3 Ways to Prevent Data Center Crises During a Power Outage

Posted on Thu, May 30, 2013
Joe Landrette
by Joe Landrette, Market Manager - Ventilation

Last year’s heavy snowstorms and Hurricane Sandy kept data center operators on their toes as they battled power outages and struggled to keep their cool – literally. These acts of nature got data center operators thinking more about ways to prevent data loss and overheating, long before the next storm hits.

1. Look to the cloud

At the Data Center World expo, from April 28 to May 2, a keynote panel featured two data center stories straight from the eye of Hurricane Sandy. For Alex Delgado, the global operations and data center manager for International Flavors & Fragrances (IFF), cloud email prevented communication headaches, and a back-up facility prevented the company from losing a single order. Donna Manley, IT senior director at the University of Pennsylvania, also relied on the cloud, backing up documentation on Box.net in case servers went down. 

2. Plan for a crisis during the design phase

For Kevin Dickens, deputy director and senior projects engineer with Jacobs Facilities, Inc., Hurricane Sandy also pointed to a need for HVAC back-up. Data center operator zColo experienced issues with its generators during the storm and decided to shut down portions of the cooling system to avoid losing the entire operation. Temperatures inside the center climbed to above 100 degrees Fahrenheit, causing zColo staff to place fans around the center, open windows and shut down some equipment.

“I have wondered what it must have been like for zColo’s mechanical team when they were informed that in order to keep IT online, they would have to shed mechanical load,” says Dickens. “Did they have any idea how the space temperature would react and how fast? Could anyone quantify the near-term and long-term risks to the IT equipment?”

Dickens addresses his own questions by pointing out that the answers should be determined during computer room air conditioner (CRAC) unit design – not during crisis time.   I would expect that the age of the CRAC’s should be reviewed along with their efficiency.  This brings us to the third topic.

3. Consider your CRAC’s age

Another consideration when creating your crisis plan: How old is the CRAC? Over 70% of the 500,000 data centers around the world are greater than 7 years old.  An old and less-efficient CRAC will put a drain on your utility bill as well as eat at your UPS capacity when crisis hits. Replacing fans within the CRACs with direct-drive EC fans means less wear and tear and, therefore, increased reliability and a longer life span.  Other benefits include:

  • 10 to 30 percent in energy efficiency gains.

  • A return on investment within six to twelve months.

  • Lowers the UPS requirements – increasing capacity for power outages.

ebm-papst offers the largest range of fan and blower sizes, along with some of the most efficient motor technology in the world.  We currently supply a range from small tubeaxial fans in blade server applications up to some of the largest fans used for data center building facilities cooling.  This proficient knowledge and product blend in IT, HVAC, and Data Center equipment position us to best address upgrades to existing cooling equipment as well as brand new cutting edge ‘green’ builds with alternative cooling methods.  

Have you recently experienced a data center crisis?  Share your story below. Interested in exploring data center cooling solutions?  Give us a call at 860-674-1515 or email sales@us.ebmpapst.com.

Tags: HVAC, Data Centers, cooling, cloud computing, CRAC