Power Capping Via Forced Idleness

We introduce a novel power capping technique, IdleCap, that achieves higher effective server frequency for a given power constraint than existing techniques. IdleCap works by repeatedly alternating between the highest performance state and a low-power idle state, maintaining a fixed average power budget, while significantly increasing the average processor frequency. In experiments conducted on an IBM BladeCenter HS21 server across three representative workloads, IdleCap reduces the mean response time by up to a factor of 3 when compared to power capping using clock-throttling. Furthermore, we argue how IdleCap applies to next-generation servers using DVFS and advanced idle states.