Sequential and Distributed Hybrid GA-SA Algorithms for Energy Optimization in Embedded Systems

Maha Idrissi Aouad (1) , Lhassane Idoumghar (2) , René Schott (3, 4) , Olivier Zendra (1)

1 TRIO - Real time and interoperability
2 LMIA - Laboratoire de Mathématiques Informatique et Applications [UHA]
3 LORIA - Laboratoire Lorrain de Recherche en Informatique et ses Applications
4 IECN - Institut Élie Cartan de Nancy

Reducing memory energy consumption in embedded systems is crucial. In this paper, we propose new hybrid sequential and distributed algorithms based on Simulated Annealing (SA) and Genetic Algorithms (GA) in order to reduce memory energy consumption in embedded systems. Our algorithms outperform the Tabu Search (TS) approach. In fact, our hybrid algorithms manage to consume nearly from 76% up to 98% less memory energy than TS. Execution time savings for the distributed version (nearly from 72% up to 74% for a cluster of 4 PCs) are also recorded.

Mots clés: Distributed systems; embedded systems; genetic algorithms; memory management; optimization; simulated annealing.; simulated annealing

Domaines: Systèmes embarqués; Intelligence artificielle [cs.AI]; Langage de programmation [cs.PL]; Optimisation et contrôle [math.OC]; Modélisation et simulation

Maha Idrissi Aouad, Lhassane Idoumghar, René Schott, Olivier Zendra. Sequential and Distributed Hybrid GA-SA Algorithms for Energy Optimization in Embedded Systems. the IADIS International Conference Applied Computing 2010, Oct 2010, Timisoara, Romania. pp.167-174. ⟨inria-00524974⟩