Malaga. Getting to/from Malaga. Cercanias train from airport.
|Title||Authors||Abstract Paper Presentation|
|A Fair Comparison of Modern CPUs and GPUs Running the Genetic Algorithm under the Knapsack Benchmark||Jiri JarosP and Petr Pospichal||abs DOI|
|Validating a Peer-to-Peer Evolutionary Algorithm||Juan Luis Jimenez LaredoP, Pascal Bouvry, Sanaz Mostaghim and Juan Julian Merelo Guervos||abs DOI slides|
|OpenCL implementation of Particle Swarm Optimization: A fair comparison between CPU and GPU performances||Stefano CagnoniP, Alessandro Bacchini, and Luca Mussi||abs DOI slides|
|FlexGP: Genetic Programming on the Cloud. Best paper||Dylan Sherry, Kalyan Veeramachaneni, James McDermott and Una-May O'ReillyP||abs DOI photo|
|Migration and Replacement Policies for Preserving Diversity in Dynamic Environments||David Millan-Ruiz and Jose Ignacio Hidalgo||abs DOI poster|
|Distributed Simulated Annealing with MapReduce||Atanas Radenski||abs DOI|
|Pool-based distributed evolutionary algorithms using an object database||Juan-J. Merelo, Antonio Mora, J. Albert-Cruz, and Anna I. Esparcia||abs DOI poster photos|
|A Library to Run Evolutionary Algorithms in the Cloud using the Hadoop MapReduce Framework||Pedro Fazenda, James McDermott and Una-May O'Reilly||abs DOI|
Each poster should fit into a 100 cm wide × 120 cm tall (approx 3'3" by 4 feet) display area.
Redcliffe Imaging can make foldable fabric posters. A cloth poster may be easier to transport to Malaga.
There is growing interest in running evolutionary computation on Parallel and Distributed Computing Infrastructures. A number of technologies are already available. These include Grid and Cloud Computing, Internet Computing (e.g. seti@home, boinc), General Purpose Computation on Graphics Processing Units (GPGPU), multi-core and many-core architectures and supercomputers. Although they are routinely used for running computing intensive applications, considerable skill is required to get the best from them. The experimenter has to consider scheduling, porting of applications, communication topologies, new parallel models and architectures, cache and memory management optimization, preemptive multitasking and simultaneous multi threading and even energy consumption. Also, the experimenter may need to change their evolutionary algorithm to fully exploit these new tools.
At EvoPar 2012 scientists and engineers gathered to share and exchange their experiences, discuss challenges, and report state-of-the-art and in-progress research on all aspects of the application of evolutionary algorithms for improving parallel architectures and distributed computing Infrastructures. EvoPAR will assist the two-way flow of ideas between the parallel computing community and the EC community.
Accepted papers are in the proceedings of Evo*, published in
7248 of the Springer Lecture Notes in Computer Science, which was
available at the Conference.
F. Fernandez de Vega, University of Extremadura, Spain
W. B. Langdon, University College London, UK
|Reviewer||Affiliation||Ignacio Hidalgo||University Complutense Madrid, Spain|
|Jose Carlos Ribeiro||Politechnique Institute of Leiria, Portugal|
|Gianluigi Folino||L'ICAR-CNR, Cosenza, Italy|
|J. J. Merelo||University of Granada, Spain|
|Leonardo Vanneschi||University of Milano-Bicocca|
|Garnett Wilson||Afinin Labs, Inc., Canada|
|Tien-Tsin Wong||The Chinese University of Hong Kong|
|Malcolm Heywood||Dalhousie University, Canada|
|Qizhi Yu||Inria, France|
|Shigeyoshi Tsutsui||Hannan University, Japan|
|Denis Robilliard||l'Universite du Littoral-Cote d'Opale, France|
|Stephane Gobron||EPFL, Switzerland|
|Ogier Maitre||Strasbourg University, France|
|Pierre Collet||Strasbourg University, France|
|Simon Harding||IDSIA, Switzerland|
|Marco Tomassini||Lausanne University, Switzerland|