Data centre dynamics may be intricate and daunting to understand, but the many building management systems (BMS) out there are carefully designed to make sense of all that information. While they all provide a comprehensive view of the data centre and its power usage, this is not worth much unless it is understood and where appropriate, acted upon.
Monitoring either remotely or locally will provide insights into a number of critical variables, be it the current, real power, apparent power, voltage, power factor or kilowatt hours. However, there are two elements to a building management system which largely relate to how this information is used – whether it creates a reactive action or a proactive reaction. There is limited reason to invest in a BMS if you fail to react to the information it reveals. While this obviously means when an alarm sounds as this will invariably suggest imminent system failure, it is equally as important to use the on-going information to constantly refine and improve how the system is operating. So, what does the reactive versus proactive approach really mean?
– The reactive enables a data centre manager to identify potential faults such as where circuits are being overloaded, which could ultimately lead to down time, equipment damage and significant loss in revenue. Typically, in these situations there will be alerts to these issues from either an individual breaker or circuit level right they way through to higher level failures. These are often quickly conveyed through alarms, emails or texts to the appropriate personnel.
– Whereas the proactive approach uses the information in a holistic way by giving detailed insights into energy usage. Interpretation and interrogation of this data provides opportunity for redefining and improving the operation to ensure maximum uptime. It can also eliminate costly responses to situations requiring instant attention, as it can identify areas which need addressing long before they become critical.
The challenge is getting the right BMS for the operation and ensuring the information revealed is acted upon. As Martin Pearce, Sales Director at Critical Power says, “it is all about getting an accessible central interface, which can operate across multiple network monitored power distribution units right the way though to branch level monitoring. From a reactive perspective this should ideally show temperature, humidity, current and unit status as a very bare minimum. But, to be able to use the information more proactively, it is more useful to have a sophisticated system which can then be used for capacity planning, maximising power circuits and calculating a data centre’s PUE and DCIE efficiency figures.”
