DCN May 2016 - Page 27

design & facilities management in a densely packed space. It may be hardware and wires, however their smooth and efficient operation is critical for day-to-day activities worldwide. One of the key concerns for data centre operators is providing consistent service availability. There can be many causes of downtime, including human error, equipment failure or external disruptions, such as changes in weather or leaks. Any outages impact on the overall performance of the data centre and subsequently on the company it provides services to. This can result in significant financial losses for both parties. If any data centre, processing support or back up system is lost for any period of time, operations of the receiving company may be disrupted or stopped. The average cost of data centre outages has now risen from over £475,000 in 2013 to around £510,000 – a 24 per cent increase over a three year period. Whereas maximum downtime costs have risen 81 per cent since 2010 to a current level of £1,665,000. Therefore, it is essential for data centre operators to ensure their infrastructure, buildings and equipment are fully protected. generators have been squeezed into spaces that were not originally designed for that purpose. Each generator requires a day tank, fuel pumps, valves, hose connections and similar fittings. Often, there is a supply and return piping connection to a larger underground diesel storage tank outside of the building footprint. An undetected diesel leak presents a major fire risk, and if ignited, can lead to a rapidly growing fire beyond the capacity of existing sprinkler systems. The fact that most data centres and server rooms have raised floors makes the risk even greater as water and fuel seek the lowest point, making raised floor facilities even more vulnerable because the leak is often concealed. Any outages impact on the overall performance of the data centre and subsequently on the company it provides services to. Feeling the heat and remains safe. It is estimated that the industry will grow from 1.58 billion square feet in 2013 to 1.94 billion square feet in 2018. To support this growth and efficiency of continuous service, it is important for data centre managers to also ensure adequate safety systems to protect vital equipment and prevent any unnecessary downtime. The impact of downtime The complex infrastructures of data centres contain thousands of optical and electrical connections concentrated Any hardware working at full capacity can easily overheat, which will have serious consequences on the overall efficiency of the data centre. In tight spaces, racked equipment generates a large amount of heat. For many facilities, HVAC equipment is the primary heat extraction system, requiring chilled water to be pumped in and out as required. However, running HVAC service pipes so close to electrical equipment poses a serious risk. In addition, some server facilities require diesel powered back up generators. In many older buildings, Cooling down It is not just heat that poses a threat to data centre safety. More an d more operators are selecting cold locations in an effort to cut cooling and power costs. From Facebook’s new data centre in Sweden to Google’s £350m data centre in Finland, companies are increasingly looking north for further expansion. This trend is driven by the fact that data centres are one of the largest and fastest growing consumers of electricity. In the US alone, data centres use an estimated 91 billion kilowatt-hours of electricity – enough to power New York City 27