An Energy-Delay Tunable Task Allocation Strategy for Collaborative Applications in Networked Embedded Systems.

[Back] [PDF]

An Energy-Delay Tunable Task Allocation Strategy for Collaborative Applications in Networked Embedded Systems
Tao Xie¹Xiao Qin²

1 Department of Computer Science
San Diego State University. San Diego, CA 92182

2 Department of Computer Science and Software Engineering

Auburn University, Auburn, AL 36849-5347.

Collaborative applications with energy and low-delay constraints are emerging in various networked embedded systems
like wireless sensor networks and multimedia terminals. Conventional energy-aware task allocation schemes developed for
collaborative applications only concentrated on energy savings when making allocation decisions. Consequently, the length of the
schedules generated by such allocation schemes could be very long, which is unfavorable or, in some situations, even not tolerated.
To remedy this problem, we developed a novel task allocation strategy called Balanced Energy-Aware Task Allocation (BEATA) for
collaborative applications running on heterogeneous networked embedded systems. The BEATA algorithm aims at blending an
energy-delay efficiency scheme with task allocations, thereby making the best trade-offs between energy savings and schedule
lengths. Aside from that, we introduced the concept of an energy-adaptive window, which is a critical parameter in the BEATA strategy.
By fine-tuning the size of the energy-adaptive window, users can readily customize BEATA to meet their specific energy-delay trade-off
needs imposed by applications. Further, we built a mathematical model to approximate the energy consumption caused by both
computation and communication activities. Experimental results show that BEATA significantly improves the performance of
embedded systems in terms of energy savings and schedule length over existing allocation schemes.
IEEE Transactions on Computers, vol. 57, no. 3, pp. 329-343, March 2008.

Acknowledgments
The work reported in this paper was supported by the US National Science Foundation under Grants No. CCF-0742187, No. CNS-0757778, No. CNS-0831502, No. OCI-0753305, and No. DUE-0621307, and Auburn University under a startup grant.