Samuel
Xavier-de-Souza

samuel at dca.ufrn.br

Dep. de Eng. de Computação e Automação
Centro de Tecnologia
Universidade Federal do Rio Grando de Norte
Av. Salgado Filho, 3000, DCA/CT/UFRN
59078-900 Natal/RN Brazil

Summary

Singular Diana's husband and proud father of three: Lena, Enio, and Anita.

Passionate about high-performance and energy-efficient computing research.

Associate Professor at Universidade Federal do Rio Grande do Norte (2009-Present).

Computer Engineering degree from Universidade Federal do Rio Grande do Norte (2000) and Doctor in Electrical Engineering degree from Katholieke Universiteit Leuven (2007), worked for IMEC (2001-2002) and for the Vlaams Supercomputer Centrum (2008).

Led and collaborated in several national and international cooperation projects, resulting in many high impact publications, including co-authorship with researchers from University of Bristol, Université de Mons, Università di Pisa, University of Queensland, École Polytechnique Fédérale de Lausanne, École Polytechnique de Montréal, and Imperial College London.

PI or co-PI of 10 funded projects, including 2 international ones, totalling over US$ 7.6M in acquired grants.

Founded in 2013 and currently leads the Laboratory of Parallel Architectures for Signal Processing, a young research lab at the Digital Metropolis Institute with dozens of researchers, among postdocs, PhD students, and MSc students under his direct supervision.

Founded in 2016 and currently directs NPAD, UFRN's high performance computing center, used by over 500 researchers resulting so far in over 160 journal publications.

Founded the startup Ycodify in 2021.

IEEE Senior Member (2019) and Royal Society-Newton Advanced Fellow (2016).

Projects

New computationally scalable methodologies for target-oriented 4D seismic in pre-salt reservoirs
Funded by Shell Brasil (Rio de Janeiro, BR)
2022-04 to 2026-03|Contract|Co-PI

GPU-CPU load balancing for high-performance simulation of wave equations for exploration geophysics
Funded by Oracle for Research (São Paulo, SP)
2021-03 to 2022-06|Award|PI

Computational Systems for Aerospace Missions (Research Network)
Funded by Universidade Federal do Rio Grande do Norte (Natal, RN)
2020-11 to 2023-10|Grant|PI

Energy-Efficient Parallel Software for Hererogeneous Processors
Funded by Universidade Federal do Rio Grande do Norte (Natal, RN)
2020-05 to 2022-04|Grant|PI

Improving the Energy Efficiency of Multi-core Virtual Machines and Software-Defined Servers
Funded by State Secretariat for Education, Research and Innovation SERI (Bern)
2018-01 to 2018-12|Grant|Co-PI

Low Frequency Multi-Core Processor for Energy-Efficient On-Board Processing in Aerospace Systems
Funded by Conselho Nacional de Desenvolvimento Científico e Tecnológico (Brasilia, DF)
2017-11 to 2020-10|Grant|PI

New Methods of Seismic Exploration by Full Waveform Inversion
Funded by Shell do Brasil (Rio de Janeiro, RJ)
2017-08 to 2022-03|Contract|Co-PI

More Power to Programmers: Enabling Energy-Effincient Software Engineering
Funded by The Royal Society (London)
2016-12-01 to 2018-12-01|Award|PI

Expansion and consolidation of UFRN's research and graduate infrastructure
Funded by Financiadora de Estudos e Projetos (Brasilia, DF)
2013-12 to 2020-12|Contract|PI

Simulation and Analysis of Algorithms for the Multi-Core Era
Funded by Conselho Nacional de Desenvolvimento Científico e Tecnológico (Brasilia, DF)
2010-11 to 2012-10|Grant|PI

Supervision

João Batista Fernandes
Ph.D. Thesis: Balanceamento de carga intra-nó em sistemas de memória distribuída heterogêneos.

Vitor Hugo Mickus Rodrigues
Ph.D. Thesis: Suporte de GPUs em Geração Automática de Códigos de Inversão Sísmica com Subdomínios.

Carla dos Santos Santana
Ph.D. Thesis: Técnicas Avançadas de Checkpointing para Inversão Sísmica.

Johannes Warwick da Costa Farias
Ph.D. Thesis: Processador Risc-V de múltiplos núcleos tolerante a falhas com mode não-seguro de alto desempenho.

Antônio Oliveira Filho
Ph.D. Thesis: Energy-Efficient Deep Learning in Graphical Processing Units.

Diego Vinícius Cirilo do Nascimento
Ph.D. Thesis: Tolerância a falhas aplicada à redução de consumo energético em processadores digitais.

Antônio Denilson de souza Oliveira
Ph.D. Thesis: Balanceador de Carga Computacional Distribuído baseado na Técnica de Work Stealing.

Maelso Bruno Pacheco Nunes Bezerra
Ph.D. Thesis: Inversão Completa de Formas de Onda em Arquiteturas de Processamento Gráfico com Linguagem Específica de Domínio.

Júlio Gustavo Soares Firmo da Costa
Ph.D. Thesis: Scalability and elasticity of low-code development driven by metadata interpretation.

Anderson Bráulio Nóbrega da Silva
Ph.D. Thesis: Energy-Efficient Scheduling of Multiple Performance-Modeled Application on Power-Modeled Architectures.

Wellington Silva de Souza
Ph.D. Thesis: Energy-Efficient Parallel Software for Data Centers.

Fabrício Costa e Silva
Ph.D. Thesis: Receptor GPS paralelo em dois passos.

Gabriel Sebastian Von Conta
Master Thesis: Integration of High-Level Domain Specific Lenguages Devito and OPS for Full Waveform Inversion.

Alumni

Ítalo Augusto Souza de Assis
Postdoctoral researcher.

Tiago Tavares Leite Barros
Postdoctoral researcher.

Lucas Costa Pereira Cavalcante
Postdoctoral researcher.

Roger Rommel Ferreira de Araújo
Ph.D. Thesis: Boosting memory access locality of the spectral element method with Hilbert Space-Filling Curves. 2021.

Demétrios Araújo Magalhães Coutinho
Ph.D. Thesis: Performance-Energy Trade-offs Prediction and Runtime Selection for Parallel Applications on Heterogeneous Multi-processing Systems. 2021.

Alex Fabiano de Araújo Furtunato
PhD Thesis: Analytical speedup modeling for parallel applications with variable memory-access delay in symmetric architectures.. 2021.

Ítalo Augusto Souza de Assis
PhD Thesis: Intra-node and Inter-node load balancing and other scalable approaches for high-performance seismic processing.. 2021.

Iria Caline Saraiva Cosme
PhD Thesis: Um método para o cálculo da inversa de matrizes em blocos com uso limitado de memória.. 2018.

Kayo Gonçalves de Silva
PhD Thesis: Órbita Perpétua e outras técnicas de otimização global robusta para os algoritmos Simulated Annealing Acoplado e Nelder-Mead.. 2018.

Carlos Avelino de Barros
PhD Thesis: Redução do consumo energético de aplicações paralelas em arquiteturas multi-core. 2016.

Reinaldo Agostinho de Souza Filho
Master Thesis: High-level programing models for RISC-V multicore processors. 2022.

Igor Macedo Silva
Master Thesis: CEVERO: um processador de hardware aberto para missões aeroespaciais. 2021.

Raffael Sadite Cordoville Gomes de Lima
Master Thesis: A parallel software-defined ultra-low-power receiver for a satellite message forwarding system. 2021.

João Batista Fernandes
Master Thesis: Granularity load scheduling auto-tuning for multi-core processors applied to reverse-time migration. 2020.

Carla dos Santos Santana
Master Thesis: Workload scheduling analysis in geophysical numerical methods. 2020.

Vitor Hugo Mickus Rodrigues
Master Thesis: Workload scheduling analysis in geophysical numerical methods. 2020.

Júlio Gustavo Soares Firmo da Costa
Master Thesis: Metadata Interpretation Driven Development, uma abordagem de desenvolvimento de Sistemas de Dissociado do Domínio de Negócios. 2020.

Victor Hugo Freitas de Oliveira
Master Thesis: Application speedup characterization: modeling parallelization overhead and variations of problem size and number of cores. 2019.

Hugo Rafael Gonçalves Cavalcante
Master Thesis: Aquisição de frequência em sinais de satélites de baixa órbita por meio do critério da máxima correntropia. 2018.

Roger Rommel Ferreira de Araújo
Master Thesis: Otimização e paralelização de um simulador de bombeio mecânico para poços direcionais. 2016.

Fabrício Costa e Silva
Master Thesis: Algoritmo eficiente para aquisição de sinais de GPS por software. 2015.

Ítalo Augusto Souza de Assis
Master Thesis: Um algoritmo paralelo eficiente de migração reversa no tempo (rtm) 3d com granularidade fina. 2015.

Demétrios Araújo Magalhães Coutinho
Master Thesis: Implementação paralela escalável e eficiente do algoritmo simplex padrão em arquitetura multicore. 2014.

Kayo Gonçalves e Silva
Master Thesis: Análise de escalabilidade de uma implementação paralela do simulated annealing acoplado. 2013.

Anderson Bráulio Nóbrega da Silva
Master Thesis: Um modelo de unidade de medida inercial utilizando três acelerômetros triaxiais. 2013.

Francisco Ary Alves de Souza
Master Thesis: Análise de desempenho da rede neural artificial do tipo multilayer perceptron na era multicore. 2012.

Desnes Augusto Nunes do Rosário
Master Thesis: Escalabilidade paralela de um algoritmo de migração reversa no tempo (RTM) pré-empilhamento. 2012.

Publications



. . p. . [bibtex]