University of Missouri-St. Louis
The increasing attention on global scheduling algorithms for identical multiprocessor platforms produced different, independently developed, schedulability tests. However, the existing relations among such tests have not been sufficiently clarified, so that it is difficult to understand which strategy provides the best performances in a particular scenario. In this paper, the authors will summarize the main existing results for the schedulability analysis of multiprocessor systems scheduled with global EDF, showing, when possible, existing dominance relations. They will compare these algorithms taking into consideration different aspects, namely, run-time complexity, and average performances over a randomly generated workload, sustainability properties and speedup factors.