As companies continue to invest in virtualization and cloud technologies, one thing is become quite clear with respect to their data centers: now more than ever, they need data center infrastructure management (DCIM) tools.
DCIM software provides a complete picture of the health of a data center and is designed to identify and resolve issues with a minimum amount of human intervention. Their ability to correlate power, cooling and space resources to individual servers means DCIM tools can proactively inform IT management systems of potential physical infrastructure problems and how they might impact specific IT loads.
Think about how virtualization affects a data center and it’s immediately clear why such DCIM capabilities are so important. For one, numerous virtual machines (VMs) and applications may be running on a single host server, making the health and availability of each server that much more critical. And many virtual environments are set up such that VMs can move from one host to another automatically, with no human intervention, based on compute requirements.
Perhaps a given application running on one VM exceeds its predefined CPU capacity, so a new VM is automatically spun up on a different physical host that has available processor capacity. While that’s great from an IT efficiency perspective, it can wreak havoc if that new host can’t handle the load from a power and cooling perspective. If the PDU providing power to the server becomes overloaded, for example, the server – and many others – could crash.
What’s required is a real-time awareness of constantly changing power and cooling capacities, so companies can make more intelligent decisions about where to install servers in the first place. As operations management tools constantly collect data from various devices in the racks, DCIM tools can use that data to perform real-time simulation of server placement based on power, cooling, space, port availability and weight requirements of the server. The tools can even generate a work order to place the server in the correct rack.
Taking such capabilities a step further, some DCIM tools can also communicate power, cooling, rack and other physical infrastructure information directly to a VM manager. That ensures VMs and the applications they support are moved only to physical hosts that have “healthy” power and cooling resources. Such capabilities not only increase IT reliability but may well reduce the need for physical infrastructure redundancy – thus saving money.
Working with VM managers, DCIM tools can save money in other ways, too. If compute demand is low at night, after work hours, virtual loads can be consolidated from two racks to one, for example. That allows the DCIM system to shut down one rack until it’s needed again, saving the energy required both to power the rack and to cool it.
With their reporting capabilities, DCIM and data center planning tools can even help IT and infrastructure teams make the case to senior management when data center upgrades are required and justified.
To learn more about how to meet the challenges that high-density computing and dynamic, virtual workloads bring, read APC by Schneider Electric white paper number 107, “How Data Center Infrastructure Management Software Improves Planning and Cuts Operational Costs.”