Data Centers Moratoriums: How to Lessen Their Impact

Increasingly, countries and large cities are having a larger say in how data centers are built and operated.

In mid-2019 the city of Amsterdam announced a year-long moratorium on new data centers. When this hiatus ended, it set out new requirements for data center projects. Among them: data centers 5MVA and up must have a power usage effectiveness of 1.2 (very efficient) and practice "nature-inclusive construction."

Singapore, a data center hub in the Asia Pacific, has just formally announced a pause in the construction of new data centers. The motivations of the island state, where data centers consume approximately 7% of the total energy used in the country, include better use of state land, the pursuit of "quality, not quantity", best-in-class resource efficiency, and a smaller carbon footprint for the sector.

Frankfurt, too, plans to put a handle on the number of new data centers in the city.

All about power and space
A common thread runs through the three cities' desire to limit and regulate the construction of new data centers: the desire to tame the sector's high demands for power and space and, in tropical Singapore's case, water. And it would be no surprise if other countries and cities were to do the same.

Given this and that there is no letting up in the volume of data that needs to be processed, how then should data center operators run their facilities such that they can forestall having to build new ones in the foreseeable future or don't find themselves having to meet new regulatory requirements in a hurry? An obvious start would be to make optimal use of data center power and space.

To optimize power and manage energy costs, operators need to first not rely on the published nameplate power ratings on their servers. While adequate in the past, this practice doesn't lend itself to accurate planning and data center management and operators need to factor in accepted industry assumptions as well.

Accurate measurement is key
Next comes putting in place the ability to measure energy efficiency, for only through individual server measurement can managers accurately know how much power their equipment is drawing and acquire precise numbers that will aid capacity and energy efficiency planning. Such monitoring can also help operators identify non-working and low-efficiency processing assets.

Other than power, there's also a need to monitor temperatures so heating, cooling and airflow can be adjusted to avoid overcooling. Tools are readily available for remote and accurate measurement and management of power, temperature and other parameters.

Optimizing space
As for space, while floor space is pretty fixed and the assets that occupy it can't be repositioned easily, opportunities can be found in the vertical plane. Ways of optimizing vertical space include using taller racks, creating more space within racks by moving network switches, patch panels and other gear outside them, and simply housekeeping the space within each rack. Upgrading to a three-phase power supply with matching power distribution units also helps to optimize space whilst boosting power capacity.

By optimizing power and space within their data centers, operators can get the maximum use of their resources while reducing their carbon footprints. Additionally, they can get more productive years out of their facilities and thus be able to put off their plans for new ones and worry less about being able to meet future regulatory requirements.