Skip to content

Capacity Management Is Key to Keeping Data Centers Running Efficiently

by on August 30, 2010

High-density hardware can save space and iron out wrinkles in the refresh cycle in data centers, but it also poses power and cooling challenges. To prevent those challenges from producing costly outages, it’s wise to have a capacity management system in place to nip problems in the bud before they bloom into trouble.

Datacenter “[M]ost data center operators are unaware of the loading and current power and cooling capability of their data centers, even at a total bulk level,” wrote Neil Rasmussen, chief innovation officer at APC-MGE, in the white paper “Power and Cooling Capacity Management for Data Centers.”

“Installing equipment that exceeds the design density of the data center, and the resultant stresses on the power and cooling systems, are causing downtime from overloads, overheating, and loss of redundancy,”
he added.

Exceeding design densities in most data centers is all too easy these days, according to Rasmussen. A fully populated rack can draw from 6kW to 35kW of power, yet most data centers are designed to accommodate 2kW per rack.

When putting together a capacity management system, data center managers should look for tools that provide real time and planning capabilities to meet the stress loads existing and future high-density equipment places on the power and cooling capacity. The best tools for managing capacity do it at the rack level. Measuring capacity at the room level is too broad to be effective, while taking the pulse of hardware at the device level produces an amount of detail that’s difficult to work with.

Another important element for a capacity management system is the ability to identify “stranded” capacity, as well as address ways to avoid it. Stranded capacity is capacity that can’t be used because other forms of capacity in the system are maxed out. Most data centers have stranded capacity issues, according to Rasmussen.

Other important components in a capacity management system include:

  • Planning for appropriate as well as sufficient capacity to prevent oversizing of the data center
  • Inclusion of capacity alarms to alert a manager when power consumption exceeds planned limits or cooling and power systems are operating at degraded levels or are unable to provide necessary redundancy
  • Ability to analyze capacity in both historical and hypothetical scenarios

Good capacity management is crucial to data center efficiency, but more improtantly, to an organization’s bottom line. If you are under utlizing your IT assets, then you are probably spending money and resources you don’t need to. If, on the other hand, you have a tightly run organization where capacity is managed effectively, then you need less IT assets to accomplish the same tasks and your overall data center effciency will increase exponentially.

What is your take on this? What are you doing in your organization to efficently utilize capacity?

Leave a Comment

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: