For simplicity, and to remain compatible with the primary scheduler, threads boosted by the primary scheduler are beyond the time slicer's control. When threads or virtual machines are boosted by virtual devices, the time slicer does not track this boosting and will consider them to be boosted to the real time priority. The time slicer continues to request the primary scheduler to schedule the next available thread at the same priority as the currently executing thread whenever a quantum expires. This allows time slicing to continue in a round robin fashion between boosted threads, but prevents the time slicer and primary scheduler from "fighting."