Power optimization of real-time embedded systems on variable speed processors

Shin, Youngsoo; Choi, Ki‐Young; Sakurai, T.

doi:10.1109/iccad.2000.896499

Cited by 43 publications

(1 citation statement)

References 15 publications

(20 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In other words, when an instance (job) of a task is assigned to a CPU, the CPU speed does not change until it is preempted or completed. This definition is somewhat different from the inter-task definition in [3,13,16,20,23] wherein the speed of each tasks is fixed and cannot be changed between different instances (jobs). Inter-task DVFS scheduling algorithms are often implemented under operating system control, and programs do not need to be modified during their runtime.…”

Section: Introductionmentioning

confidence: 99%

Time and Energy Efficient DVS Scheduling for Real-Time Pinwheel Tasks

Chen

2014

Journal of Applied Research and Technology

View full text Add to dashboard Cite

Dynamic voltage/frequency scaling (DVFS) is one of the most effective techniques for reducing energy use. In this paper, we focus on the pinwheel task model to develop a variable voltage processor with d discrete voltage/speed levels. Depending on the granularity of execution unit to which voltage scaling is applied, DVFS scheduling can be defined in two categories: (i) inter-task DVFS and (ii) intra-task DVFS. In the periodic pinwheel task model, we modified the definitions of both intra-and inter-task and design their DVFS scheduling to reduce the power consumption of DVFS processors. Many previous approaches have solved DVFS problems by generating a canonical schedule in advance and thus require pseudo polynomial time and space because the length of a canonical schedule depends on the hyperperiod of the task periods and is generally of exponential length. To limit the length of the canonical schedules and predict their task execution, tasks with arbitrary periods are first transformed into harmonic periods and their key features are profiled. The proposed methods have polynomial time and space complexities, and experimental results show that, under identical assumptions, the proposed methods achieve more energy savings than the previous methods.

show abstract