2026-07-23T06:16:22Z https://drops.dagstuhl.de/oai

oai:drops-oai.dagstuhl.de:16330 2024-03-06T10:57:40Z ddc:004 open_access

Response-Time Analysis for Self-Suspending Tasks Under EDF Scheduling Aromolo, Federico Biondi, Alessandro Nelissen, Geoffrey Real-Time Systems Schedulability Analysis Self-Suspending Tasks EDF Scheduling The self-suspending task model proved to be particularly effective in capturing the timing behavior of real-time systems characterized by complex execution patterns, such as computation offloading to hardware accelerators, inter-core synchronization by means of multiprocessor locking protocols, and highly parallel computation. Most of the existing results for the timing analysis of self-suspending tasks do not support the widely adopted Earliest Deadline First (EDF) scheduling algorithm, being instead primarily focused on fixed-priority scheduling. This paper presents a response-time analysis for constrained-deadline self-suspending tasks scheduled under EDF on a uniprocessor system. The proposed analysis is based on a model transformation from self-suspending sporadic tasks to sporadic tasks with jitter, which can then be analyzed using a state-of-the-art analysis method for EDF scheduling. Experimental results are presented to compare the performance of the proposed technique in terms of schedulability ratio with that of the pessimistic suspension-oblivious approach and with a less general technique for task sets with implicit deadlines. Schloss Dagstuhl – Leibniz-Zentrum für Informatik Federico Aromolo and Alessandro Biondi and Geoffrey Nelissen 2022 Is Part Of LIPIcs, Volume 231, 34th Euromicro Conference on Real-Time Systems (ECRTS 2022) InProceedings Text doc-type:ResearchArticle publishedVersion application/pdf doi:10.4230/LIPIcs.ECRTS.2022.13 urn:nbn:de:0030-drops-163306 https://drops.dagstuhl.de/entities/document/10.4230/LIPIcs.ECRTS.2022.13 eng https://creativecommons.org/licenses/by/4.0/legalcode