High Performance Computing for Computational Science – VECPAR 2018: 13th International Conference, São Pedro, Brazil, September 17-19, 2018, Revised Selected Papers

Description

This book constitutes the thoroughly refereed post-conference proceedings of the 13th International Conference on High Performance Computing in Computational Science, VECPAR 2018, held in São Pedro, Brazil, in September 2018. The 17 full papers and one short paper included in this book were carefully reviewed and selected from 32 submissions presented at the conference. The papers cover the following topics: heterogeneous systems, shared memory systems and GPUs, and techniques including domain decomposition, scheduling and load balancing, with a strong focus on computational science applications. Front Matter ....Pages i-xiii Front Matter ....Pages 1-1 Communication–Free Parallel Mesh Multiplication for Large Scale Simulations (Rômulo M. Silva, Benaia S. J. Lima, José J. Camata, Renato N. Elias, Alvaro L. G. A. Coutinho)....Pages 3-15 Dynamic Configuration of CUDA Runtime Variables for CDP-Based Divide-and-Conquer Algorithms (Tiago Carneiro, Jan Gmys, Nouredine Melab, Francisco Heron de Carvalho Junior, Pedro Pedrosa Rebouças Filho, Daniel Tuyttens)....Pages 16-30 Design, Implementation and Performance Analysis of a CFD Task-Based Application for Heterogeneous CPU/GPU Resources (Lucas Leandro Nesi, Lucas Mello Schnorr, Philippe Olivier Alexandre Navaux)....Pages 31-44 Optimizing Packed String Matching on AVX2 Platform (Mehmet Akif Aydoğmuş, M. Oğuzhan Külekci)....Pages 45-61 A GPU-Based Metaheuristic for Workflow Scheduling on Clouds (Elliod Cieza, Luan Teylo, Yuri Frota, Cristiana Bentes, Lúcia M. A. Drummond)....Pages 62-76 A Systematic Mapping on High-Performance Computing for Protein Structure Prediction (Gesiel Rios Lopes, Paulo Sergio Lopes de Souza, Alexandre C. B. Delbem)....Pages 77-91 Performance Evaluation of Deep Learning Frameworks over Different Architectures (Rafael Gauna Trindade, João Vicente Ferreira Lima, Andrea Schwerner Charão)....Pages 92-104 Non-uniform Domain Decomposition for Heterogeneous Accelerated Processing Units (Gabriel Freytag, Philippe Olivier Alexandre Navaux, João Vicente Ferreira Lima, Lucas Mello Schnorr, Paolo Rech)....Pages 105-118 Performance Evaluation of Two Load Balancing Algorithms for Hybrid Clusters (Tiago Marques do Nascimento, Rodrigo Weber dos Santos, Marcelo Lobosco)....Pages 119-131 An Improved OpenMP Implementation of the TVD–Hopmoc Method Based on a Cluster of Points (Frederico Cabral, Carla Osthoff, Roberto Pinto Souto, Gabriel P. Costa, Sanderson L. Gonzaga de Oliveira, Diego N. Brandão et al.)....Pages 132-145 A Scheduling Theory Framework for GPU Tasks Efficient Execution (Antonio-Jose Lázaro-Muñoz, Bernabé López-Albelda, Jose María González-Linares, Nicolás Guil)....Pages 146-159 A Timer-Augmented Cost Function for Load Balanced DSMC (William McDoniel, Paolo Bientinesi)....Pages 160-173 Accelerating Scientific Applications on Heterogeneous Systems with HybridOMP (Matthias Diener, Daniel J. Bodony, Laxmikant Kale)....Pages 174-187 A New Parallel Benchmark for Performance Evaluation and Energy Consumption (Adriano Marques Garcia, Claudio Schepke, Alessandro Gonçalves Girardi, Sherlon Almeida da Silva)....Pages 188-201 Bigger Buffer k-d Trees on Multi-Many-Core Systems (Fabian Gieseke, Cosmin Eugen Oancea, Ashish Mahabal, Christian Igel, Tom Heskes)....Pages 202-214 A Parallel Generator of Non-Hermitian Matrices Computed from Given Spectra (Xinzhe Wu, Serge G. Petiton, Yutong Lu)....Pages 215-229 LRMalloc: A Modern and Competitive Lock-Free Dynamic Memory Allocator (Ricardo Leite, Ricardo Rocha)....Pages 230-243 Front Matter ....Pages 245-245 Towards a Strategy for Performance Prediction on Heterogeneous Architectures (Silvio Stanzani, Raphael Cóbe, Jefferson Fialho, Rogério Iope, Marco Gomes, Artur Baruchi et al.)....Pages 247-253 Back Matter ....Pages 255-264