TY - JOUR
T1 - Memory-centric scheduling for multicore hard real-time systems
AU - Yao, Gang
AU - Pellizzoni, Rodolfo
AU - Bak, Stanley
AU - Betti, Emiliano
AU - Caccamo, Marco
N1 - Funding Information:
Acknowledgements The authors would like to thank Lui Sha for his insightful suggestions. The material presented in this paper is based upon work supported by Lockheed Martin, NSERC and NSF under Award No. CNS-1035736. Any opinions, findings, and conclusions or recommendations expressed in this publication are those of the authors and do not necessarily reflect the views of the Lockheed Martin, NSERC or NSF.
PY - 2012/11
Y1 - 2012/11
N2 - Memory resources are a serious bottleneck in many real-time multicore systems. Previous work has shown that, in the worst case, execution time of memory intensive tasks can grow linearly with the number of cores in the system. To improve hard real-time utilization, a real-time multicore system should be scheduled according to a memory-centric scheduling approach if its workload is dominated by memory intensive tasks. In this work, a memory-centric scheduling technique is proposed where (a) core isolation is provided through a coarse-grained (high-level) Time Division Multiple Access (TDMA) memory schedule; and (b) the scheduling policy of each core "promotes" the priority of its memory intensive computations above CPU-only computation when memory access is permitted by the high-level schedule. Our evaluation reveals that under high memory demand, our scheduling approach can improve hard real-time task utilization significantly compared to traditional multicore scheduling.
AB - Memory resources are a serious bottleneck in many real-time multicore systems. Previous work has shown that, in the worst case, execution time of memory intensive tasks can grow linearly with the number of cores in the system. To improve hard real-time utilization, a real-time multicore system should be scheduled according to a memory-centric scheduling approach if its workload is dominated by memory intensive tasks. In this work, a memory-centric scheduling technique is proposed where (a) core isolation is provided through a coarse-grained (high-level) Time Division Multiple Access (TDMA) memory schedule; and (b) the scheduling policy of each core "promotes" the priority of its memory intensive computations above CPU-only computation when memory access is permitted by the high-level schedule. Our evaluation reveals that under high memory demand, our scheduling approach can improve hard real-time task utilization significantly compared to traditional multicore scheduling.
KW - CPU memory co-scheduling
KW - Hard real-time system
KW - Memory-centric scheduling
KW - Multi-core scheduling
UR - http://www.scopus.com/inward/record.url?scp=84869096276&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=84869096276&partnerID=8YFLogxK
U2 - 10.1007/s11241-012-9158-9
DO - 10.1007/s11241-012-9158-9
M3 - Article
AN - SCOPUS:84869096276
SN - 0922-6443
VL - 48
SP - 681
EP - 715
JO - Real-Time Systems
JF - Real-Time Systems
IS - 6
ER -