Technology with the Environment in Mind

  Section 3 of 11  

Dynamic Data Center Power Management: Trends, Issues, and Solutions

CURRENT POWER ALLOCATION METHODS

A typical data center power distribution hierarchy is designed to deliver a fixed amount of power to the room and then to each rack. The challenge of the data center operator is to determine the number of servers for each rack while ensuring that the overall rack (hence room) power consumption does not exceed the limit. To do this, the operator must make certain assumptions about the maximum power consumption per server. For most data centers, there are two ways of determining this: 1) using server nameplate power value, and 2) using a derated nameplate value.

The server nameplate value, which is marked on the server by the manufacturer, is the maximum possible power value that the server can consume. Actual power consumption is typically much less that the nameplate power. Most data center operators are aware that typical server power consumption never reaches the nameplate value, and one way for them to increase server density is to derate the nameplate power by a certain percentage—depending on the workload that is deployed on the server.

While derating, as opposed to using nameplate value, can improve server density, it is obvious that both methods are not optimal. The result is the following:

1. Under-utilization of available power for computing: A static allocation of power based on the worst-case scenario planning leads to inefficiencies and does not maximize the use of available power capacity.
2. Under-population of rack space: This is a direct result of the static allocation approach described above. The wastage of rack space is more severe when racks are populated using the nameplate power.
3. The two factors above result in higher energy costs than actually necessary, primarily from the cost of cooling the room that is sparsely populated with servers.
4. Unnecessary power cooling capacity expansion: Although existing power capacity is not being utilized effectively (as described in 1, 2, and 3 above), if the data center needs to deploy new services, the operator has no choice but to expand power/cooling capacity or even build a new data center at very high costs.

From the analysis above, we can see that the current power allocation approach results in wastage and high total cost of ownership. A better approach is needed.

  Section 3 of 11  

Back to Top