Saw this over at the AMD blog recently, certainly sounds pretty neat, the ability to control the power usage of your CPU on a per-watt basis, which is significantly more effective than current power limiting strategies available today.
Well, let’s say that you have a maximum power draw on your fully configured server of 300W, and you have 42 slots in your server. The simple math says that you have 12.6Kw of power load that you need to be able to support. Now, if your power budget only allows you to bring 12Kw to the rack, you essentially have 2 slots that need to be left open in the rack because you can only support 40 and not 42 servers. But, by utilizing a custom TDP, you could drop the max power that some servers could draw, bringing you in under the limit of 12Kw and still getting 42 servers in the rack.
No mention on whether or not the processor will turn off cores once you hit certain thresholds (say you reduce a 80W 12-core cpu to 60W of power, it may be better to run fewer cores with higher clock speeds than more cores on lower clock speeds depending on the workload).