Hybrid algorithms for dynamic schedulability testing
Charlie McElhone, Alan Burns, Robert Ian Davis · 2002
The paper describes an investigation into the performance of a variety of algorithms for dynamic schedulability testing. A real time node may run its own set of resident periodic tasks and may also receive aperiodic requests for computations to be completed within hard deadlines. In order to decide whether each request is schedulable, and can therefore be accepted, the node may use a guarantee algorithm. Such algorithms trade off complexity against pessimism. The paper describes how algorithms of hybrid complexity can achieve the best trade off for given characteristics of the resident periodic task set.