TY - JOUR
T1 - Optimizing resource speed for two-stage real-time tasks
AU - Melani, Alessandra
AU - Mancuso, Renato
AU - Cullina, Daniel
AU - Caccamo, Marco
AU - Thiele, Lothar
N1 - Funding Information:
The material presented in this paper is based upon work supported by the National Science Foundation (NSF) under Grant Numbers CNS-1219064 and CNS-1302563. 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 NSF.
Publisher Copyright:
© 2016, Springer Science+Business Media New York.
PY - 2017/1/1
Y1 - 2017/1/1
N2 - Multiple resource co-scheduling algorithms and pipelined execution models are becoming increasingly popular, as they better capture the heterogeneous nature of modern architectures. The problem of scheduling tasks composed of multiple stages tied to different resources goes under the name of “flow-shop scheduling”. This problem, studied since the ‘50s to optimize production plants, is known to be NP-hard in the general case. In this paper, we consider a specific instance of the flow-shop task model that captures the behavior of a two-resource (DMA-CPU) system. In this setting, we study the problem of selecting the optimal operating speed of the two resources with the goal of minimizing power usage while meeting real-time schedulability constraints. In particular, we derive an algorithm that finds the optimal speed of one resource while the speed of the other resource is kept constant. Then, we discuss how to extend the proposed approach to jointly optimize the speed of the two resources. In addition, applications to multiprocessor systems and energy minimization are considered. All the proposed algorithms run in polynomial time, hence they are suitable for online operation even in the presence of variable real-time workload.
AB - Multiple resource co-scheduling algorithms and pipelined execution models are becoming increasingly popular, as they better capture the heterogeneous nature of modern architectures. The problem of scheduling tasks composed of multiple stages tied to different resources goes under the name of “flow-shop scheduling”. This problem, studied since the ‘50s to optimize production plants, is known to be NP-hard in the general case. In this paper, we consider a specific instance of the flow-shop task model that captures the behavior of a two-resource (DMA-CPU) system. In this setting, we study the problem of selecting the optimal operating speed of the two resources with the goal of minimizing power usage while meeting real-time schedulability constraints. In particular, we derive an algorithm that finds the optimal speed of one resource while the speed of the other resource is kept constant. Then, we discuss how to extend the proposed approach to jointly optimize the speed of the two resources. In addition, applications to multiprocessor systems and energy minimization are considered. All the proposed algorithms run in polynomial time, hence they are suitable for online operation even in the presence of variable real-time workload.
UR - http://www.scopus.com/inward/record.url?scp=84988698736&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=84988698736&partnerID=8YFLogxK
U2 - 10.1007/s11241-016-9259-y
DO - 10.1007/s11241-016-9259-y
M3 - Article
AN - SCOPUS:84988698736
SN - 0922-6443
VL - 53
SP - 82
EP - 120
JO - Real-Time Systems
JF - Real-Time Systems
IS - 1
ER -