Abstract

The problem of mapping tasks onto a computational grid with the aim to minimize the power consumption and the makespan subject to the constraints of deadlines and architectural requirements is considered in this paper. To solve this problem, we propose a solution from cooperative game theory based on the concept of Nash Bargaining Solution. The proposed game theoretical technique is compared against several traditional techniques. The experimental results show that when the deadline constraints are tight, the proposed technique achieves superior performance and reports competitive performance relative to the optimal solution. measures. Power management can be achieved by two methods. The Dynamic Power Management (DPM) (27) approach brings a processor into a power-down mode, where only certain parts of the computer system (e.g., clock generation and time circuits) are kept running, while the processor is in an idle state. The Dynamic Voltage Scaling (DVS) (34) approach exploits the convex relation between the CPU supply voltage and power consumption. The rationale behind DVS technique is to stretch out task execution time through CPU frequency and voltage reduction. Power-aware resource allocation using DVS can be classified as static and dynamic techniques. Static techniques are applied at design time by allocating and scheduling resources using off-line approaches, while dynamic techniques control the runtime behavior of the systems to reduce power consumption. The traditional resource allocation and scheduling theory deals with fixed CPU speed, and hence cannot be directly applied to this situation. In this paper, we study the problem of power-aware task allocation (PATA) for assigning a set of tasks onto a computational grid each equipped with DVS feature. The PATA problem is formulated as multi- constrained multi-objective extension of the Generalized Assignment Problem (GAP). PATA is then solved using a novel solution from cooperative game theory based on the celebrated Nash Bargaining Solution (NBS) (22); we shall acronym this solution concept as NBS-PATA. The rest of the paper is organized as follows: A brief discussion of related work is presented in Section II. The PATA problem formulation and background information are discussed in Section III. In Section IV, we model a cooperative game played among the machines for task allocation with the objective to minimize power consumption and makespan, simultaneously. Experimental results and concluding remarks are provided in Sections V and VI, respectively.

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call

Disclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.