David Pereira
Leonel Sousa
ABSTRACT
In this paper, a novel game theory based approach for task scheduling on emerging heterogeneous embedded systems is proposed. It relies on the auction concept to assign tasks to players, where players compete against each other by bidding for the tasks in order to acquire them. To ensure the feasibility of game rounds, a set of different utility functions are formulated, such that each player can select its own best strategy, which corresponds to the frequency level that guarantees the minimum energy consumption variation at the overall system level. The proposed energy-aware multiplayer auction-based scheduling approach was extensively evaluated across a set of 13 real applications from 4 standard benchmark suites on the state-of-the-art ARM Juno r2 platform with two different multi-core clusters as composite players, i.e, dual-core Cortex-A72 and a quad-core Cortex-A53 clusters. For different benchmark combinations, the experimental results show that the proposed approach allows achieving significant energy savings when compared to the standard ARM Linaro, GTS and EAS approaches, i.e., up to 36%, 32% and 22%, respectively.
- ARM. ARM Versatile Express Juno r2 Development Platform Technical Reference Manual, 2015.Google Scholar
- N. Bielik et al. Cooperative versus non-cooperative game theoretical techniques for energy aware task scheduling. In Intl. Green Comp. Conf., page 6, 2012. Google ScholarDigital Library
- F. Gaspar et al. A framework for application guided task management on heterogeneous embedded systems. ACM Trans. on Arch. and Code Opt., 12(4):1--25, 2016. Google ScholarDigital Library
- T. S. Muthukaruppan et al. Price Theory Based Power Management for Heterogeneous Multi-Cores. In ASPLOS, pages 161--176, 2014.Google ScholarDigital Library
- R. B. Myerson. Game Theory: Analysis of Conflict. Harvard University Press, 1997.Google Scholar
- D. Pereira. Boosting Energy-Efficiency of Heterogeneous Systems via Game Theory. Master's thesis, IST, Universidade de Lisboa, Portugal, 2016.Google Scholar
- D. Puschini et al. A Game-Theoretic Approach for Run-Time Distributed Optimization on MP-SoC. Intl. J. of Reconfigurable Computing, 2008(403086):11, 2008.Google Scholar
- J. Wilkins et al. Optimizing performance and energy in computational grids using non-cooperative game theory. In Intl. Green Comp. Conf., pages 343--355, 2010. Google ScholarDigital Library
- G. Wu et al. An Energy-Aware Multi-Core Scheduler based on Generalized Tit-For-Tat Cooperative Game. Journal of Computers, 7(1):106--115, 2012.Google ScholarCross Ref
Recommendations
Load-matching adaptive task scheduling for energy efficiency in energy harvesting real-time embedded systems
ISLPED '10: Proceedings of the 16th ACM/IEEE international symposium on Low power electronics and designIn this paper we present a load matching task scheduling algorithm for energy harvesting real-time embedded systems using a realistic model for the battery charging and discharging processes. The proposed approach addresses two important issues that ...
Discovering theorems in game theory: Two-person games with unique pure Nash equilibrium payoffs
In this paper we provide a logical framework for two-person finite games in strategic form, and use it to design a computer program for discovering some classes of games that have unique pure Nash equilibrium payoffs. The classes of games that we ...
Bounded energy allocation and scheduling for real-time Embedded Systems
Real-Time and Embedded Computing SystemsFor energy-constrained real-time embedded systems, the power-delay tradeoff property of Volta ge-Clock Scaling (VCS) needs to be carefully considered in scheduling real-time tasks for meeting strict timing requirements. In addition, non-real-time tasks ...
Comments