Energy-aware load balancing of parallel evolutionary algorithms with heavy fitness functions in heterogeneous CPU-GPU architectures.
Juan José EscobarJulio Ortega LoperaAntonio Francisco DíazJesús GonzálezMiguel DamasPublished in: Concurr. Comput. Pract. Exp. (2019)
Keyphrases
- evolutionary algorithm
- load balancing
- energy aware
- graphics processing units
- fitness function
- high performance computing
- grid computing
- fault tolerance
- dynamic load balancing
- grid environment
- parallel computing
- genetic programming
- pc cluster
- load balancing strategy
- massively parallel
- load balance
- fault tolerant
- peer to peer
- parallel processing
- parallel implementation
- differential evolution
- computing resources
- genetic algorithm
- distributed systems
- energy consumption
- computing systems
- wireless sensor networks
- data skew
- routing algorithm
- distributed computing
- mobile agents
- data center
- smart home
- routing protocol
- differential evolution algorithm
- initial population
- sensor networks
- real time
- computing environments
- response time
- data transfer
- multithreading
- computational intelligence
- message passing