Energy-aware Supply Voltage and Body Biasing Voltage Scheduling Algorithm for Real-time Distributed Systems  

在线阅读下载全文

作  者:SUYajuan WEIShaojun 

机构地区:[1]InstituteofMicroelectronics,TsinghuaUniversity,Beijing100084China

出  处:《Chinese Journal of Electronics》2005年第2期259-263,共5页电子学报(英文版)

摘  要:Technique of energy minimization by combining Dynamic voltage scheduling (DVS) and Adaptive body biasing voltage (ABB) method for distributed realtime system at design level is proposed. First, a simplified energy optimizing model is illustrated where the supply voltage or body biasing voltage is kept as constant according to each separated frequency region, thus calculation of exceeding equation is avoided. Divergence of simplified and analytic model within 5% indicates the accuracy of this model. Based on it, the proposed approach named LEVVS (Low energy supply voltage and body biasing voltage scheduling algorithm) explores space of minimizing energy consumption by finding optimal trade-off between dynamic and static energy. The corresponding optimal supply voltage and body biasing voltage are determined by an iterative method in which the supply voltage and body biasing voltage of tasks are adjusted according to the value of energy latency differential coefficient of each task and slack time distribution of the system. Experiments show that using LEVVS approach, 51% more average energy reduction can be obtained than employing DVS method alone. Furthermore the effects of switch capacitance and global slack on the energy saving efficiency of LEVVS are investigated. The smaller the global slack or average switch capacitance is, the more the energy saving of LEVVS compared with DVS is.

关 键 词:电压调度 LEVVS 实时分布系统 能量微分系数 

分 类 号:TM73[电气工程—电力系统及自动化]

 

参考文献:

正在载入数据...

 

二级参考文献:

正在载入数据...

 

耦合文献:

正在载入数据...

 

引证文献:

正在载入数据...

 

二级引证文献:

正在载入数据...

 

同被引文献:

正在载入数据...

 

相关期刊文献:

正在载入数据...

相关的主题
相关的作者对象
相关的机构对象