By Topic

Energy-Aware Scheduling Algorithm for Task Execution Cycles with Normal Distribution on Heterogeneous Computing Systems

Sign In

Cookies must be enabled to login.After enabling cookies , please use refresh or reload or ctrl+f5 on the browser for the login options.

Formats Non-Member Member
$31 $13
Learn how you can qualify for the best price for this item!
Become an IEEE Member or Subscribe to
IEEE Xplore for exclusive pricing!
close button

puzzle piece

IEEE membership options for an individual and IEEE Xplore subscriptions for an organization offer the most affordable access to essential journal articles, conference papers, standards, eBooks, and eLearning courses.

Learn more about:

IEEE membership

IEEE Xplore subscriptions

3 Author(s)
Kenli Li ; Nat. Supercomput. Center in Changsha, Hunan Univ., Changsha, China ; Xiaoyong Tang ; Qifeng Yin

In the past few years, many energy-aware scheduling algorithms have been developed primarily using the dynamic voltage-frequency scaling (DVFS) capability which has been incorporated into recent commodity processors. However, these techniques are unsatisfied with optimizing both schedule length and energy consumption. Furthermore, most algorithms schedule tasks according to their average case execution time and not consider the task's execution cycles with probability distribution in real-world. In recognition of this, we study the problem of scheduling independent stochastic tasks with normal distribution, deadline and energy consumption budget constraints on a heterogeneous platform. We first formulate this energy-aware stochastic scheduling problem as a linear programming, which maximize the guaranteed confidence probabilities under deadline and energy consumption budget constraints. Then, we propose a heuristic energy-aware stochastic tasks scheduling algorithm (ESTS) to solve this problem, which can achieve high schedule performance for independent tasks with lower complexity. Our extensive simulation performance evaluation study, based on randomly generated stochastic applications and real-world applications, clearly demonstrate that our proposed heuristic algorithm can improve system guaranteed confidence probability and has a good trade-off between schedule length and energy consumption.

Published in:

Parallel Processing (ICPP), 2012 41st International Conference on

Date of Conference:

10-13 Sept. 2012