Learning-Directed Dynamic Voltage and Frequency Scaling Scheme with Adjustable Performance for Single-Core and Multi-Core Embedded and Mobile Systems
Dynamic voltage and frequency scaling (DVFS) is a well-known method for saving energy consumption. Several DVFS studies have applied learning-based methods to implement the DVFS prediction model instead of complicated mathematical models. This paper proposes a lightweight learning-directed DVFS meth...
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2018-09-01
|
| Series: | Sensors |
| Subjects: | |
| Online Access: | http://www.mdpi.com/1424-8220/18/9/3068 |
| _version_ | 1811307281076191232 |
|---|---|
| author | Yen-Lin Chen Ming-Feng Chang Chao-Wei Yu Xiu-Zhi Chen Wen-Yew Liang |
| author_facet | Yen-Lin Chen Ming-Feng Chang Chao-Wei Yu Xiu-Zhi Chen Wen-Yew Liang |
| author_sort | Yen-Lin Chen |
| collection | DOAJ |
| description | Dynamic voltage and frequency scaling (DVFS) is a well-known method for saving energy consumption. Several DVFS studies have applied learning-based methods to implement the DVFS prediction model instead of complicated mathematical models. This paper proposes a lightweight learning-directed DVFS method that involves using counter propagation networks to sense and classify the task behavior and predict the best voltage/frequency setting for the system. An intelligent adjustment mechanism for performance is also provided to users under various performance requirements. The comparative experimental results of the proposed algorithms and other competitive techniques are evaluated on the NVIDIA JETSON Tegra K1 multicore platform and Intel PXA270 embedded platforms. The results demonstrate that the learning-directed DVFS method can accurately predict the suitable central processing unit (CPU) frequency, given the runtime statistical information of a running program, and achieve an energy savings rate up to 42%. Through this method, users can easily achieve effective energy consumption and performance by specifying the factors of performance loss. |
| first_indexed | 2024-04-13T09:01:45Z |
| format | Article |
| id | doaj.art-07f4efc03e4444d9b0548730ceed2799 |
| institution | Directory Open Access Journal |
| issn | 1424-8220 |
| language | English |
| last_indexed | 2024-04-13T09:01:45Z |
| publishDate | 2018-09-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Sensors |
| spelling | doaj.art-07f4efc03e4444d9b0548730ceed27992022-12-22T02:53:07ZengMDPI AGSensors1424-82202018-09-01189306810.3390/s18093068s18093068Learning-Directed Dynamic Voltage and Frequency Scaling Scheme with Adjustable Performance for Single-Core and Multi-Core Embedded and Mobile SystemsYen-Lin Chen0Ming-Feng Chang1Chao-Wei Yu2Xiu-Zhi Chen3Wen-Yew Liang4Department of Computer Science and Information Engineering, National Taipei University of Technology, Taipei 10608, TaiwanMediaTek Inc., Hsinchu 30078, TaiwanDepartment of Computer Science and Information Engineering, National Taipei University of Technology, Taipei 10608, TaiwanDepartment of Computer Science and Information Engineering, National Taipei University of Technology, Taipei 10608, TaiwanDepartment of Computer Science and Information Engineering, National Taipei University of Technology, Taipei 10608, TaiwanDynamic voltage and frequency scaling (DVFS) is a well-known method for saving energy consumption. Several DVFS studies have applied learning-based methods to implement the DVFS prediction model instead of complicated mathematical models. This paper proposes a lightweight learning-directed DVFS method that involves using counter propagation networks to sense and classify the task behavior and predict the best voltage/frequency setting for the system. An intelligent adjustment mechanism for performance is also provided to users under various performance requirements. The comparative experimental results of the proposed algorithms and other competitive techniques are evaluated on the NVIDIA JETSON Tegra K1 multicore platform and Intel PXA270 embedded platforms. The results demonstrate that the learning-directed DVFS method can accurately predict the suitable central processing unit (CPU) frequency, given the runtime statistical information of a running program, and achieve an energy savings rate up to 42%. Through this method, users can easily achieve effective energy consumption and performance by specifying the factors of performance loss.http://www.mdpi.com/1424-8220/18/9/3068dynamic voltage and frequency scaling (DVFS)embedded systemsenergy consumptionlow-power software designmulticore computing systemsmobile devices |
| spellingShingle | Yen-Lin Chen Ming-Feng Chang Chao-Wei Yu Xiu-Zhi Chen Wen-Yew Liang Learning-Directed Dynamic Voltage and Frequency Scaling Scheme with Adjustable Performance for Single-Core and Multi-Core Embedded and Mobile Systems Sensors dynamic voltage and frequency scaling (DVFS) embedded systems energy consumption low-power software design multicore computing systems mobile devices |
| title | Learning-Directed Dynamic Voltage and Frequency Scaling Scheme with Adjustable Performance for Single-Core and Multi-Core Embedded and Mobile Systems |
| title_full | Learning-Directed Dynamic Voltage and Frequency Scaling Scheme with Adjustable Performance for Single-Core and Multi-Core Embedded and Mobile Systems |
| title_fullStr | Learning-Directed Dynamic Voltage and Frequency Scaling Scheme with Adjustable Performance for Single-Core and Multi-Core Embedded and Mobile Systems |
| title_full_unstemmed | Learning-Directed Dynamic Voltage and Frequency Scaling Scheme with Adjustable Performance for Single-Core and Multi-Core Embedded and Mobile Systems |
| title_short | Learning-Directed Dynamic Voltage and Frequency Scaling Scheme with Adjustable Performance for Single-Core and Multi-Core Embedded and Mobile Systems |
| title_sort | learning directed dynamic voltage and frequency scaling scheme with adjustable performance for single core and multi core embedded and mobile systems |
| topic | dynamic voltage and frequency scaling (DVFS) embedded systems energy consumption low-power software design multicore computing systems mobile devices |
| url | http://www.mdpi.com/1424-8220/18/9/3068 |
| work_keys_str_mv | AT yenlinchen learningdirecteddynamicvoltageandfrequencyscalingschemewithadjustableperformanceforsinglecoreandmulticoreembeddedandmobilesystems AT mingfengchang learningdirecteddynamicvoltageandfrequencyscalingschemewithadjustableperformanceforsinglecoreandmulticoreembeddedandmobilesystems AT chaoweiyu learningdirecteddynamicvoltageandfrequencyscalingschemewithadjustableperformanceforsinglecoreandmulticoreembeddedandmobilesystems AT xiuzhichen learningdirecteddynamicvoltageandfrequencyscalingschemewithadjustableperformanceforsinglecoreandmulticoreembeddedandmobilesystems AT wenyewliang learningdirecteddynamicvoltageandfrequencyscalingschemewithadjustableperformanceforsinglecoreandmulticoreembeddedandmobilesystems |