Abstract
Original language  Undefined 

Supervisors/Advisors 

Sponsors  
Date of Award  18 Jun 2014 
Place of Publication  Enschede 
Publisher  
Print ISBNs  9789036536790 
DOIs  
State  Published  18 Jun 2014 
Fingerprint
Keywords
 Real time systems
 EWI24855
 Power Management
 Constrained optimisation
 NWO 612.063.715
Cite this
}
Algorithmic power management  Energy minimisation under realtime constraints. / Gerards, Marco Egbertus Theodorus.
Enschede : Centre for Telematics and Information Technology (CTIT), 2014. 162 p.Research output: Scientific › PhD Thesis  Research UT, graduation UT
TY  THES
T1  Algorithmic power management  Energy minimisation under realtime constraints
AU  Gerards,Marco Egbertus Theodorus
N1  NWO 612.063.715
PY  2014/6/18
Y1  2014/6/18
N2  Energy consumption is a major concern for designers of embedded devices. Especially for battery operated systems (like many embedded systems), the energy consumption limits the time for which a device can be active, and the amount of processing that can take place. In this thesis we study how the energy consumption can be reduced for certain classes of realtime applications. To minimise the energy consumption, we introduce several algorithms that combine power management techniques with scheduling (algorithmic power management). The power management techniques that we focus on are speed scaling and sleep modes. When the processor (or some peripheral) is active, its speed, and with it the supply voltage, can be decreased to reduce the power consumption (speed scaling), while when the processor is idle it can be put in a low power mode (sleep modes). The resulting problem is to determine a schedule, speeds for the processors (which may vary over time) and/or times when a device is put to sleep. We discuss energy minimisation for three classes of realtime systems, namely (1) realtime systems with agreeable deadlines, (2) realtime systems with precedence constraints, and (3) framebased realtime systems. Below we elaborate on these classes of realtime systems. (1) For realtime systems with agreeable deadlines it holds that an earlier arrival time implies an earlier deadline (and vice versa). Compared to existing methods our algorithms can reduce the energy consumption by up to 54% for the considered multimedia workloads, and our evaluation shows that these algorithms are near optimal even with inaccurate predictions. (2) The second class of realtime systems we focus on are tasks with precedence constraints that must be scheduled on a multicore system and for which the speeds have to be determined. We derive a scheduling criterion that implicitly assigns speeds and minimises the energy consumption. (3) In the third setting, we study the optimal combination of speed scaling, sleep modes and scheduling for framebased realtime systems. While the literature considers only trivial schedules for this problem, we study energy optimal schedules for such systems.
AB  Energy consumption is a major concern for designers of embedded devices. Especially for battery operated systems (like many embedded systems), the energy consumption limits the time for which a device can be active, and the amount of processing that can take place. In this thesis we study how the energy consumption can be reduced for certain classes of realtime applications. To minimise the energy consumption, we introduce several algorithms that combine power management techniques with scheduling (algorithmic power management). The power management techniques that we focus on are speed scaling and sleep modes. When the processor (or some peripheral) is active, its speed, and with it the supply voltage, can be decreased to reduce the power consumption (speed scaling), while when the processor is idle it can be put in a low power mode (sleep modes). The resulting problem is to determine a schedule, speeds for the processors (which may vary over time) and/or times when a device is put to sleep. We discuss energy minimisation for three classes of realtime systems, namely (1) realtime systems with agreeable deadlines, (2) realtime systems with precedence constraints, and (3) framebased realtime systems. Below we elaborate on these classes of realtime systems. (1) For realtime systems with agreeable deadlines it holds that an earlier arrival time implies an earlier deadline (and vice versa). Compared to existing methods our algorithms can reduce the energy consumption by up to 54% for the considered multimedia workloads, and our evaluation shows that these algorithms are near optimal even with inaccurate predictions. (2) The second class of realtime systems we focus on are tasks with precedence constraints that must be scheduled on a multicore system and for which the speeds have to be determined. We derive a scheduling criterion that implicitly assigns speeds and minimises the energy consumption. (3) In the third setting, we study the optimal combination of speed scaling, sleep modes and scheduling for framebased realtime systems. While the literature considers only trivial schedules for this problem, we study energy optimal schedules for such systems.
KW  Real time systems
KW  EWI24855
KW  Power Management
KW  Constrained optimisation
KW  NWO 612.063.715
U2  10.3990/1.9789036536790
DO  10.3990/1.9789036536790
M3  PhD Thesis  Research UT, graduation UT
SN  9789036536790
PB  Centre for Telematics and Information Technology (CTIT)
ER 