Organizations globally are looking to get their data centers tier certified. While many understand the objective of getting certified according to a certain data center tier classification, others
Data center tier certifications focus only the components of a data center that directly affect uptime. The four data center tier classifications articulate the performance of the data center and are aimed to give organizations the assurance that their data centers have attained the desired uptime and performance. However, there are a lot of misconceptions regarding what a particular data center tier classification requires.
Myth 1: Data center tier certifications define the quality levels
Data center tier classifications are not aimed at defining a data center as good, better or best. They indicate that data centers certified according to different data center tier classifications are designed to meet varied business needs and performance requirements. They only help organizations figure what needs to be done for the data center to meet the business need.
For instance, a university would have performance requirements different from a healthcare company. Most of the data centers that have been certified by us as tier 4 belong to a bank or a financial service company. These organizations require levels of uptime or availability that are also mandated by the tier 4 data center tier classification.
Some co-location providers want to get their data centers tier 4 certified in the quest for a competitive edge. However, they don’t realize the cost and complexity associated in reaching that level.
Myth 2: Data center tier certification mandates multiple service providers
The tier certifications do not require a data center to have power supply from multiple service providers. Thus, a data center that is certified as tier 3 or tier 4 can have power supply from only one utility service provider. In fact, we have also carried out data center tier certification for a data center located in a very remote area, with no utility service provider.
The Uptime Institute only looks at parts of the data center infrastructure, controlled by the owner. In this case, the utility service provider cannot be controlled by the owner. Therefore, we look at the engine backup generator as the primary source of reliable power.
While a data center can have power from a single utility service provider, the organization must ensure that there are enough redundant backup generators to provide the required levels of uptime in case the power supply fails.
Myth 3: A data center can be certified for a fractional or aggregate tier
Some organizations declare their data centers as tier 3- (minus) or tier 2.7 certified. We do not recognize such data centers as certified for any particular data center tier classification. We do not certify a data center for a tier if it meets only certain specifications or almost meets the tier certification requirements.
Myth 4: Tier certifications require that data center server rooms have raised flooring
The data center tier certifications do not mandate a single way of design or implementation for a data center and encourage owner preferences. They are based on the fundamental concepts of designing a data center and allow the owner or the designer to arrive at a solution that meets these standards. Thus, a data center owner can have his own uninterrupted power supply system, cooling system or backup generator as long as it meets the necessary uptime requirements.
The data center tier certification process only demonstrates that the data center has been designed according to the fundamental concepts and meets the necessary uptime or availability requirements.
About the authors: Julian Kudritzki is the Vice President and Vince Renaud is the Managing Principal for Uptime Institute.
(As told to Harshal Kallyanpur)
This was first published in April 2011