In order to determine how different allocations of processing power affect the mean response time we assume that allocation decisions and processor reallocations can be performed instantaneously with negligible overhead. This is not unreasonable since we are only comparing dynamic scheduling algorithms (i.e., we are not comparing with static policies). Furthermore, experimental studies have demonstrated that although overhead is required to reallocate processors, the overhead does not significantly impact dynamic scheduling algorithms [25, 24]. For the purposes of simplifying the implementation of our simulator, we also assume that the number of processors allocated to a job may be fractional. Such an assumption does not affect the qualitative results of this work. As well, the effects of this assumption on the quantitative results are insignificant, since a large number of processors are used. The simulation results reported in this paper all assume a multiprocessor system containing P = 100 processors. For implementation purposes, the algorithms studied here could be easily modified to use an integral number of processors.