A multiprocessor implementation of the total bandwidth server

Summary form only given. If a periodic task system is scheduled upon an identical multiprocessor platform using the earliest deadline first scheduling algorithm, it is known that the "schedulable utilization" - the largest bound such that any periodic task system with cumulative utilization no larger than this bound is guaranteed to be successfully scheduled - is strictly less than the capacity of the platform. The issue of using the excess processing capacity (the difference between the platform capacity and the schedulable utilization) is addressed here, and an algorithm is presented, and proven correct, that uses this excess capacity to provide guaranteed real-time service to aperiodic jobs.