For most companies, data center capacity comes at a premium, so it’s important to be able to wring every bit of it out of the space and infrastructure you have in place. But unless you have a proper strategy and tools in place for power for space, power and cooling capacity management, chances are good (100%?) that you’re not getting the most out of what you have.
One issue that affects many data centers is stranded capacity, which is data center capacity that can’t be used by IT loads due to a lack of one or more of the following resources:
- Floor and rack space
- Power distribution
- Cooling distribution
To function properly, any given IT device needs all five of those elements in sufficient quantity. In reality, the elements are almost never delivered in the ideal quantity to match the IT load. But it’s when things really get out of whack that you wind up with stranded capacity, and it can happen in many ways.
Perhaps your air conditioner has sufficient capacity but the air distribution isn’t effective enough to get the cool air to the IT load. Maybe you’ve got a PDU that has spare capacity but no available breaker positions. Or maybe you’ve got open floor space where there’s no available power or, conversely, power capacity where there’s no floor space. The list goes on and on.
Correcting the issues that lead to stranded capacity always comes at a cost, the only question is how much. Maybe you can get away with some relatively minor reconfiguration but it’s often necessary to perform major reconstruction or install new power and cooling components.
An effective capacity management system not only identifies and highlights stranded capacity, but also helps you avoid creating it in the first place. You can try to create a capacity management system by taking measurements that technicians take and scribbling them on paper or pouring them into spreadsheets, but it will quickly become outdated. As data centers become ever more virtualized, with loads moving dynamically from one server to another, you really need networked power and cooling instrumentation feeding data center infrastructure management (DCIM) software.
Such a system will present you with capacity data, showing current supply and demand conditions, including any stranded capacity. The system will also enable you to set a capacity plan, ideally during the initial design of the data center, and will alert you to any violations of that plan. It will also enable you to model any proposed changes, so you can identify problem areas ahead of time.
DCIM software will almost always include capacity planning and management tools that help you visualize, track and ultimately avoid stranded capacity of all types. Schneider Electric’s StruxureWare for Data Centers DCIM suite is one such example. To learn about these tools, visit http://www.apc.com/struxureware/us/en
To learn more about how to avoid stranded capacity, read APC by Schneider Electric white paper number 150, “Power and Cooling Capacity Management for Data Centers.”