Subscribe / Unsubscribe Enewsletters | Login | Register

Pencil Banner

Addressing power and cooling issues in virtualised data centres

Jack Loo | April 3, 2012
If left unresolved, there will be additional overhead costs from idle power and ineffective cooling capacity.

By deploying a row-based cooling system that is instrumented to automatically detect and respond to temperature changes, cooling can be provided where and when needed, only in the amount needed, thus increasing efficiency of the virtualisation .

Underloading of power and cooling systems

Since virtualisation can significantly reduce load, an oversized cooling and power infrastructure can be an important efficiency issue in a virtualised data centre. Due to the presence of fixed losses, when server consolidation and server power management significantly reduce IT load, efficiency will decrease if the power and cooling system stays the same. While the electric bill will indeed go down because of the lower IT load and less air conditioning needed to cool it, the proportion of utility power that reaches the IT loads will drop, which signifies wasted power that could be conserved further by reducing energy consumption.

Virtualisation offers a new chance for businesses to take advantage of scalable power and cooling systems. Power and cooling devices that can scale in capacity will reduce fixed losses and increase efficiency. Scalable architecture will facilitate not only downsizing to follow IT consolidation, but also subsequent re-growth to follow expansion of the now-virtualised IT load. This will allow businesses to achieve not only higher efficiency and lower energy cost, but also a data centre with the capability to meet further business demands.

Need for rack-level, real-time management of capacities

The dynamic nature of virtualised computing demands accurate, timely and actionable information about power and cooling capacities to ensure that power and cooling are keeping up with a changing load profile at the row rack and server level.

Effective deployment of a total virtualisation solution requires a system-level view and the shift towards virtualisation, with its new challenges for physical infrastructure, re-emphasises the need for integrated solutions using a holistic approach. Demands and capacities must be monitored coordinated and managed by a central system, in real time, to ensure efficient use of resources and to warn of scarce or unusable ones. Capacity management provides instrumentation and software for real-time monitoring and analysis of information about power, cooling and physical space, the three essential capacities of data centres.

3. What are the differences in power and cooling in a new data centre versus an existing data centre?

For new data centres, a capacity plan must be established during the design of the data centre. With today's scalable data centre power and cooling solutions, it is possible for new data centres to have a capacity plan that can adapt to IT growth plans in order to optimise capacity expense and electrical efficiency. 

It's important to not just ensure sufficient capacity, but also ensure the appropriate amount of capacity. Too often the focus for new data centre is on assuring sufficient capacity without regard for right-sizing to the actual IT needs. The common result is over-sizing with the associated waste of capital expenditures and electrical efficiency.


Previous Page  1  2  3  Next Page 

Sign up for Computerworld eNewsletters.