Title :
A Discrete Dynamic Voltage and Frequency Scaling Algorithm Based on Task Graph Unrolling for Multiprocessor System
Author :
Liu, Shaobo ; Qiu, Meikang
Author_Institution :
Dept. of Electr. & Comput. Eng., SUNY - Binghamton Univ., Binghamton, NY, USA
Abstract :
In this paper, we propose a two-phase discrete DVFS algorithm for a multiprocessor system, dedicated to applications where the period of task graph is smaller than the deadline of task. In those applications, multiple task graphs are unrolled and then concatenated together to form a new task graph. The proposed DVFS algorithm is applied to the new-formed task graph to stretch tasks´ execution time, lower the operating frequencies of processor and achieve the system power efficiency. Experimental results show that the proposed algorithm reduces the energy dissipation by 25% on average, comparing to previous DVFS approaches.
Keywords :
graph theory; multiprocessing systems; power aware computing; discrete dynamic voltage and frequency scaling algorithm; energy dissipation; multiple task graphs; multiprocessor system; system power efficiency; task graph unrolling; two-phase discrete DVFS algorithm; Dynamic voltage scaling; Embedded computing; Energy consumption; Energy dissipation; Feeds; Frequency; Multiprocessing systems; Telephone sets; Timing; USA Councils;
Conference_Titel :
Scalable Computing and Communications; Eighth International Conference on Embedded Computing, 2009. SCALCOM-EMBEDDEDCOM'09. International Conference on
Conference_Location :
Dalian
Print_ISBN :
978-0-7695-3825-9
DOI :
10.1109/EmbeddedCom-ScalCom.2009.11