Title :
Federated Scheduling of Sporadic DAG Task Systems
Author_Institution :
Univ. of North Carolina at Chapel Hill, Chapel Hill, NC, USA
Abstract :
Recurrent (periodic and sporadic) task systems have traditionally been scheduled upon multiprocessor platforms using either the partitioned or the global approach. Under the recently-proposed federated approach, each task is either restricted to execute upon a single processor (as in partitioned scheduling), or may execute upon multiple processors but is the only task to execute upon each of these processors. Earlier studies concerning the federated scheduling of task systems represented using the sporadic DAG model were restricted to implicit-deadline and constrained-deadline sporadic task systems, the research reported here extends this study to the consideration of task systems represented using the more general arbitrary-deadline sporadic DAG model.
Keywords :
computational complexity; directed graphs; multiprocessing systems; processor scheduling; arbitrary-deadline sporadic DAG model; computational complexity; constrained-deadline sporadic task systems; directed acyclic graph; federated scheduling; implicit-deadline sporadic task systems; multiple processors; multiprocessor platforms; partitioned scheduling; periodic task systems; recurrent task systems; single processor; sporadic DAG task systems; Analytical models; Computational modeling; Real-time systems; Scheduling; Scheduling algorithms;
Conference_Titel :
Parallel and Distributed Processing Symposium (IPDPS), 2015 IEEE International
Conference_Location :
Hyderabad
DOI :
10.1109/IPDPS.2015.33