Schneider Electric, a specialist in energy management, has accumulated a wealth of knowledge, research and expertise in providing high-density, scalable and efficient data centre physical infrastructure for virtualised environments, which are the foundation of cloud computing.
“Fact is, the current cloud computing business model offers all the computing power you require as you need it,” says Eben Owen, E&S sales manager at Schneider Electric South Africa. He however cautions IT professionals who embrace the cloud model that they need to ensure that the power and cooling physical infrastructure of their organisation is cloud-ready first.
“This is especially true in terms of density and scalability in order to deliver predictable, high efficiency operation,” he says.
High density implies the use of fewer metres of floor space with fewer, more powerful servers. Since more power is concentrated in each rack, cooling becomes more of a challenge if traditional pre-cloud cooling systems are expected to shoulder the heat load.
Owen also advises that the cloud data centre infrastructure must be more flexible because, as server utilisation is optimised via software, loads move from place to place in the data centre. The physical infrastructure must constantly adapt to the IT load through dynamic power and cooling infrastructure.
“Without the assistance of robust management software, this most certainly becomes too complex for humans to manage,” says Owen.
Predictive management tools are a safe way to maintain an efficient, lean data centre in a cloud environment.
“Tools such as Schneider Electric’s StruxureWare for Data Centres suite work in real time with VM managers (such as VMware vSphere and Microsoft System Centre Virtual Machine Manager) to assure that power and cooling capacity can efficiently accommodate the dynamic loads,” says Owen.
Cloud data centres are operated with a high level of IT fault-tolerance and this may reduce the need for power and cooling infrastructure redundancy. If, for example, the failure of a particular uninterruptible power supply (UPS) does not result in business disruption, a backup, redundant UPS system may not be needed for the one that just failed.
Matching of physical infrastructure redundancy to the fault-tolerant nature of a virtualised cloud environment is another form of the rightsizing. Rightsizing in this way can further reduce energy consumption, capital costs and fixed losses, all while improving the data centre power usage effectiveness (PUE).
“An important first step in achieving the goal of a reliable, efficient and flexible data centre, is being aware of the challenges that cloud computing can impose on data centre physical infrastructure,” says Owen.