DCN March 2017 | Page 24

software & applications
able to identify waste heat to be recycled and used in other parts of the facility as a result of identifying areas of inefficiency .
Using solid software monitoring techniques it becomes possible to change the way that the end user behaves by identifying power hungry applications and working to reduce inefficient processes . By having this data at their disposal IT managers can provide greater value to customers .
Data centre predictive modelling
reliability , and / or they may have misperceptions about the tradeoff between energy efficiency and performance . The cost of downtime is one of the driving factors in the hesitancies of deploying new monitoring and managing solutions . A recent study estimated the average downtime cost at $ 5,000 per minute , meaning that IT managers are averse to taking risks and , as a result , build in more power redundancy than necessary or overcool the data centre by setting the temperature too low and the air volume too high . These cautious approaches are adding to the inefficiency of their facilities .
Monitoring software works hand-in-hand with energy capping controls which empower data centre managers to allocate server density secure in the knowledge that it will not draw excessive power . When load spikes draw more energy , power capping controls can prevent damage to hardware and reduce the likelihood of expensive outages . When an outage does occur in emergency
Back in 2014 , US data centres alone consumed an estimated 91 billion kilowatt-hours of electricity .
situations the increased usage visibility that results from good monitoring systems enables the data centre manager to allocate resources with knowledge of which areas can pick up the slack .
The DCIM must be combined with a process for notification , escalation and resolution to be effective . The use of infrastructure standards for the uniform identification and management of equipment throughout the following systems : Electrical distribution , heating , ventilation and air conditioning ( HVAC ), connectivity and access , fire and life safety is essential .
Ventilation and cooling is another area which greatly benefits from data analytics . The seasonal variation in conditions create a challenge for data centre managers who want to optimise their outgoings on climate control . Predictive tools can be utilised by having a broad set of data that tracks data centre conditions in relation to the outside environment . Some innovative data centre operators have been
The advent of data centre predictive modelling ( DCPM ) is allowing for planners to better understand energy efficiency before the construction of the data centre has even begun . DCPM can bring together a multitude of variables , such as processor types , servers and other hardware as well as the types of workloads and power profile of a facility in order to predictively model the expected energy efficiency of a future data centre . Through this type of technology , it is possible to experiment more freely and increase the likelihood of achieving efficiency goals .
The demand for data is only going to increase . As data centres grow , a smart approach to investment informed by good efficiency models created through performance monitoring is far more effective than simply spending on large infrastructure overhauls . The call for greater energy efficiency does not simply come from a need to reduce the company bottom line , it is a global environmental need . Software monitoring means that companies can share their efficiency insights , and in turn work collaboratively towards data centres that have a much smaller environmental impact on our world .
24 | March 2017