Back to Search Start Over

Energy-Aware Real-Time Task Scheduling in Multiprocessor Systems Using a Hybrid Genetic Algorithm.

Authors :
Mahmood, Amjad
Khan, Salman A.
Albalooshi, Fawzi
Awwad, Noor
Source :
Electronics (2079-9292); 2017, Vol. 6 Issue 2, p40, 17p
Publication Year :
2017

Abstract

Minimizing power consumption to prolong battery life has become an important design issue for portable battery-operated devices such as smartphones and personal digital assistants (PDAs). On a Dynamic Voltage Scaling (DVS) enabled processor, power consumption can be reduced by scaling down the operating frequency of the processor whenever the full processing speed is not required. Real-time task scheduling is a complex and challenging problem for DVS-enabled multiprocessor systems. This paper first formulates the real-time task scheduling for DVS-enabled multiprocessor systems as a combinatorial optimization problem. It then proposes a genetic algorithm that is hybridized with the stochastic evolution algorithm to allocate and schedule real-time tasks with precedence constraints. It presents specialized crossover and perturb operations as well as a topology preserving algorithm to generate the initial population. A comprehensive simulation study has been done using synthetic and real benchmark data to evaluate the performance of the proposed Hybrid Genetic Algorithm (HGA) in terms of solution quality and efficiency. The performance of the proposed HGA has been compared with the genetic algorithm, particle swarm optimization, cuckoo search, and ant colony optimization. The simulation results show that HGA outperforms the other algorithms in terms of solution quality. [ABSTRACT FROM AUTHOR]

Details

Language :
English
ISSN :
20799292
Volume :
6
Issue :
2
Database :
Complementary Index
Journal :
Electronics (2079-9292)
Publication Type :
Academic Journal
Accession number :
123799787
Full Text :
https://doi.org/10.3390/electronics6020040